In practical applications, input saturation is an unavoidable phenomenon that can degrade system performance and even lead to instability. This work, therefore, addresses the practical finite-time formation control problem of two-wheeled mobile robots under input saturation, a consequence of the physical limitations of their drive motors. To mitigate the adverse effects of input saturation while achieving finite-time convergence, a saturated controller is developed to realize leader-follower formation control. Furthermore, event-triggered strategies—both with and without continuous monitoring—are incorporated into the proposed controller to accommodate the limited computational resources of two-wheeled mobile robots. The Zeno phenomenon is explicitly excluded by ensuring that all inter-event intervals are bounded below by a strictly positive constant. The experimental results demonstrate the effectiveness of the developed controller in achieving finite-time formation for two-wheeled mobile robots in real-world applications.

This article proposes a novel model-based planning framework for freeway ramp metering (RM), denoted as Koopman-driven linearized model-based offline planning (KLMOP). This framework integrates the model predictive control (MPC) and offline reinforcement learning (RL) under assumptions of a linear Markov decision process (MDP) with the Koopman operator. KLMOP introduces a fully linearized control framework by learning and modeling the dynamics, reward function, and value function in a latent space through a Koopman-based latent dynamical model (KLDM) and a pessimistic value iteration (PEVI) algorithm. This formulation builds upon the connection between Koopman operator theory and linear MDP. Contrastive learning is employed to ensure the expressiveness and structural conditions of the latent representation in linear MDP, enabling accurate reward prediction and efficient policy optimization. The MPC-based planning policy, then, leverages these components to solve a linear MPC problem efficiently in the latent space. Extensive simulation studies demonstrate that KLMOP significantly improves computational efficiency and control performance as compared with existing baseline methods for RM control. This framework provides a theoretically grounded and computationally efficient approach to linearizing nonlinear control problems, and its learning-based design makes it adaptable to broader applications.

This article proposes a sampled-data event-triggered adaptive neural network (NN) control strategy to cope with the digital communication and attack compensation problems of networked systems with actuator attacks and exogenous disturbance. By combining event triggering state, parameter estimation signals, and disturbance observer, a novel digital state feedback controller is designed to reduce its updating frequency and compensate for the deliberate impact of unknown actuator attacks. Moreover, considering that the state is partially measurable, a novel observer-based digital controller is designed via a double-ended event-triggering mechanism (ETM). Then, two new Lyapunov functionals are created to analyze the system stability, and two design methods are given to solve the control gain. Finally, the feasibility and validity of the derived results are verified by a visual servo control system and an offshore structure system.

While the Vision Transformer (ViT) architecture gains prominence in computer vision and finds growing applications in edge computing, its lack of strong inductive biases regarding shift, scale, and rotational invariance necessitates pre-training on large-scale datasets. Moreover, the increasing depth and parameter counts in ViT models present significant challenges for training, particularly in edge environments where computational resources are constrained. To mitigate these challenges, this paper introduces a novel Horizontally Scalable Vision Transformer (HSViT) architecture. Specifically, a novel image-level feature embedding approach is introduced that incorporates convolutional layers prior to the Transformer blocks. This design helps preserve inductive biases, allowing the model to potentially eliminate the need for pre-training while achieving strong performance on small datasets. Furthermore, a novel horizontally scalable architecture is designed, facilitating collaborative model training and inference across multiple edge devices. The experimental results show that, without pre-training, HSViT achieves up to 10% higher top-1 accuracy than state-of-the-art methods on several small datasets, while improving the top-1 accuracy of existing CNN backbones by up to 3.1% on ImageNet-1k. The code is available at https://github.com/xuchenhao001/HSViT.

Motion cueing algorithm aims to replicate realistic motion sensations for drivers while adhering to the physical limitations of the simulation platform. Model Predictive Control has been extensively employed within the domain of motion cueing algorithms for vehicle and flight simulators due to its ability to handle system constraints and optimise motion fidelity. However, traditional Model Predictive Control-based motion cueing algorithm rely on manually tuned cost function weights, which can be suboptimal and difficult to determine for different operating conditions. This suboptimal tuning can cause discrepancies between the visual input perceived by the simulator driver and the motion cues processed by their vestibular system, potentially resulting in motion sensation errors and increased risk of motion sickness. In this paper, we propose a reinforcement learning weight optimisation approach for the model predictive control-based motion cueing algorithm, leveraging Deep Q-Networks to determine an optimal set of cost function weights through training in a simulated environment. The optimised weights aim to minimise the cost function, thereby maximising the reward function. Simulation results indicate that the proposed method outperforms the traditional approach, leading to a reduction in motion sensation errors between the simulator and real vehicle driver and improving platform utilisation. The reinforcement learning-based control achieves better correlation between the reference and simulated signals for both sensed specific force and angular velocity, enhancing overall motion fidelity. It also reduces the root mean square error for sensed specific force, ensuring more accurate replication of target motion cues. Additionally, the method enables broader use of the simulator's linear displacement range, confirming the effectiveness of reinforcement learning in tuning control parameters for superior simulation performance.

DUO-Net is proposed, a unified multi-task learning framework for joint 2D object detection and depth estimation. The architecture employs a shared ResNet-based backbone with attention modules and task-specific heads to simultaneously perform bounding box localisation and dense depth prediction. A two-stage training method is adopted to sequentially pretrain each task and subsequently refine them through joint learning, enhancing both features and convergence reliability. To address task imbalance and noisy supervision, we incorporate uncertainty-aware loss weighting, enabling the model to dynamically adjust task contributions during training. Evaluated on the KITTI and JRDB datasets, DUO-Net demonstrates robust performance across both tasks while maintaining efficiency and scalability.

While deep learning models have seen significant success across various domains, their black-box learning nature and lack of interpretability affect their reliability in safety-critical applications like medical diagnostics and autonomous vehicles. In an attempt to address these limitations, Bayesian neural networks (BNNs) offer a promising alternative by incorporating uncertainty estimation into model predictions, enhancing transparency and decision-making. However, BNN development has primarily focused on efficient, high-fidelity approximate inference and guaranteed convergence in asymptotic settings. These are unsuitable for modern high-dimensional, multi-modal, and non-asymptotic deep learning applications, undermining their theoretical advantages. To bridge this gap, this paper provides in-depth reviews on how approximate Bayesian inference leverages deep learning optimization to achieve high efficiency and fidelity in high-dimensional spaces and multi-modal loss landscapes. It also reconciles Bayesian consistency with generalization objectives in non-asymptotic settings and investigates the generalization capabilities of BNNs. Additionally, this survey examines the often-overlooked expressiveness of BNNs, emphasizing how weight uncertainty and the absence of in-between uncertainty affect their performance. This survey aims to inspire BNN practitioners to adopt a deep learning perspective and offer valuable insights to propel further advancements in the field.

Automated patient monitoring solutions greatly benefit from audio emotion classification, although the considerable variance in individual expression and interpretation of emotions poses a challenge. Current approaches often employ standard Audio Spectrogram Transformer (AST) and deep learning models such as Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN)-based networks. However, their performance can be enhanced by integrating neural architecture search techniques using swarm optimisation algorithms. In this research, we explore AST with hyperparameter optimisation for speech emotion recognition. Three deep learning architectures with optimisable τ b -block structures and variable filter numbers, i.e. 1DCNN, bidirectional LSTM (BiLSTM) and CNN-BiLSTM, are also proposed, enabling the optimisation of network depth and width. A novel Cluster Search Optimisation (CSO) algorithm is introduced. It incorporates Cluster Centroid Search, a Cluster Distance Improvement metric and reinforcement learning to dispatch different search actions based on clustering convergence and Q -learning strategies, respectively. A novel Noise Tempered K-means (NTKM) clustering model is also proposed with the integration of Gaussian-based noise insertion and cluster compactness-separation measurement, to further fine-tune the cluster centriods obtained using OPTICS clustering. CSO is used for hyperparameter and architecture search for AST and aforementioned deep networks. Attention mechanisms are also integrated with CSO-optimised networks to further enhance feature learning. We evaluate the resulting models against those devised by other optimisation algorithms across the EMO-DB, SAVEE, and TESS datasets. The empirical results demonstrate that CSO-optimised AST and CNN-BiLSTM with attention mechanisms outperform other architectures and yield favourable comparison results against those from existing state-of-the-art audio emotion classification methods. • Evolving transformer and deep networks are devised for audio emotion recognition. • A Cluster Search Optimisation algorithm is proposed to adapt hyperparameters. • It incorporates Noise Tempered K-means clustering and Cluster Distance Improvement. • The Q-learning algorithm is used to optimise search behaviours. • Our study indicates CSO-optimised deep networks’ effectiveness across datasets.

Abstract Trajectory tracking with disturbance rejection is a challenging problem in robotics, particularly in applications involving selective compliance articulated robot arms (SCARA). In this paper, we address the trajectory tracking problem with the presence of disturbances in applying SCARA, by designing controllers with active force control (AFC)-based control methods. AFC has shown potential in disturbance rejection, and its per efficiency of the designed controllers, we integrated different machine learning techniques into the AFC controller, including iterative learning (IL), adaptive neuro-fuzzy inference system (ANFIS) and reinforcement learning (RL). Two case studies were conducted and compared with two different benchmark controllers to validate intelligent AFC-based controllers: a port-controlled Hamiltonian (PCH) control and a hybrid proportional-integral-derivative (PID) control. The results demonstrate that the AFC-based controllers consistently outperform the benchmark methods. Specifically, in Case 1, the AFC-RL controller achieves a 99.99% improvement in root mean square error for joint 1 compared to the hybrid PID control. In Case 2, the AFC-RL controller outperforms the AFC-IL controller in trajectory tracking accuracy by 98.71%. Also, disturbance rejection ability was tested on the AFC-based controllers with various types of disturbances. Among the three AFC-based controllers, AFC-RL shows the best performance. The findings highlight the potential of integrating machine learning into AFC for more accurate and efficient robotic control.

Abstract Traffic accidents are a major global health and economic concern. As such, research into understanding driving behaviors becomes essential to minimize the associated risks. Among various factors that can influence driving behaviors, listening to music while driving is a complex task that needs investigation. Music can enhance arousal and manage stress, which could potentially improve one’s driving performance. Listening to music, however, also competes for cognitive resources, increasing one’s mental workload and potentially degrading the driving ability. This study investigates the impact of listening to music with different tempos on drivers and their performance using an Artificial Intelligence (AI) approach. A total of 26 participants are subjected to three driving scenarios in a unique simulated experiment, while utilizing a motion platform. The conditions are driving while listening to slow-tempo music, and fast-tempo music, and with no music. A dataset is created by collecting data through Tobii eye-tracking glasses, Equivital sensor belts, and a software tool. This dataset is preprocessed and used to train a convolutional neural network-long short-term memory (CNN-LSTM) model. This model’s performance is optimized through hyperparameter tuning and Chi-squared feature selection, in order to maximize accuracy and minimize computation time. The model performance is compared with those from a densely layered deep learning model and several classical machine learning models. The devised CNN-LSTM model outperforms other machine learning models, achieving an average accuracy rate of 99.29% with minimal variance across multiple evaluations, demonstrating its effectiveness and consistency in classifying drivers’ behaviors under varying auditory conditions.

Video Instance Segmentation (VIS), also known as multi-object tracking and segmentation, represents a fundamental challenge in computer vision that requires simultaneous detection, segmentation, and tracking of object instances across video frames. This complex task has gained significant attention due to its crucial role in various real-world applications. The advent of deep learning has promoted VIS approaches, leading to numerous architectural innovations and performance improvements. This survey presents a systematic review of deep learning-based VIS methods, introducing a novel categorization based on temporal modeling strategies: frame-by-frame, clip-based, in-memory feature propagation, and in-memory object query propagation. Comprehensive quantitative comparisons of existing work across three major VIS benchmark datasets are also provided. Additionally, emerging challenges in the field are explored, with several promising research directions identified, aiming to provide valuable insights for researchers and practitioners interested in VIS, while further advancing deep learning techniques for VIS. • Categorization of VIS approaches based on their temporal modeling strategies. • Comprehensive quantitative comparison of current VIS methods. • Analysis of the challenges and potential future research directions in VIS.

Abstract For many unmanned aerial vehicle (UAV)-based applications, especially those that need to operate with resource-limited edge networked devices in real-time, it is crucial to have a lightweight computing model for data processing and analysis. In this study, we focus on UAV-based forest fire imagery detection using a lightweight convolution neural network (CNN). The task is challenging owing to complex image backgrounds and insufficient training samples. Specifically, we enhance the MobileNetV2 model with an attention mechanism for UAV-based image classification. The proposed model first employs a transfer learning strategy that leverages the pre-trained weights from ImageNet to expedite learning. Then, the model incorporates randomly initialised weights and dropout mechanisms to mitigate over-fitting during training. In addition, an ensemble framework with a majority voting scheme is adopted to improve the classification performance. A case study on forest fire scenes classification with benchmark and real-world images is demonstrated. The results on a publicly available UAV-based image data set reveal the competitiveness of our proposed model as compared with those from existing methods. In addition, based on a set of self-collected images with complex backgrounds, the proposed model illustrates its generalisation capability to undertake forest fire classification tasks with aerial images.

The extraction of fine-grained spatial-temporal characteristics for emotion classification is a challenging task owing to the subtlety and ambiguity of emotional expressions through video and audio channels. In this research, we propose an audio-visual ensemble model, comprising a two-stream 3D Convolutional Neural Network (CNN) architecture with RGB and optical flow as inputs for video emotion classification, as well as a variant of Wav2Vec2 for audio emotion recognition. The Wav2Vec2 variant integrates additional recurrent and attention layers with each transformer block to extract long- and short-term dependencies. A new Particle Swarm Optimisation (PSO) algorithm is proposed to fine-tune hyper-parameters of 3D CNNs and the enhanced Wav2Vec2, and formulate audio-visual ensemble models with the smallest sizes. It integrates a reinforcement learning (RL) algorithm, i.e. Asynchronous Advantage Actor-Critic (A3C), for search parameter and hybrid leader construction, and another RL algorithm, Proximal Policy Optimisation (PPO), for search action selection, as well as hypotrochoid and super formula-based search operations. Evaluated using audio-visual emotion datasets, our evolving ensemble model outperforms those devised by other search methods and existing state-of-the-art deep networks, significantly.

In this paper, the reliable tracking control and collision avoidance problems for cyber-physical systems (CPSs) with false data injection (FDI) attacks are investigated. FDI attacks can significantly compromise the safety and performance of CPSs by corrupting control and navigation data. Safety is an important aspect of CPSs. Unmanned ground vehicles and aerial vehicles are important applications of CPSs. The dual challenge of maintaining system safety and stability under deliberate cyberattacks, while ensuring reliable obstacle avoidance in dynamic environments, remains unresolved in many current methodologies. These challenges are amplified in CPSs owing to their reliance on real-time data and their susceptibility to adversarial manipulation. The main objective of this study is to develop a resilient tracking control strategy that can effectively mitigate the impact of FDI attacks and achieve obstacle avoidance. We propose a novel framework based on the exponential control barrier function (ECBF) and a novel observer-based auxiliary signal approach that can ensure the resilience of CPSs against FDI attacks and obstacle avoidance. The effectiveness of our methods is shown through extensive simulations and physical experiments, which depict improved tracking accuracy and system stability in the presence of FDI attacks as compared with those of traditional control methods.Note to Practitioners—This research tackles the problem of maintaining dependable tracking control and avoiding collisions in CPSs vulnerable to FDI attacks, with applications in fields such as smart transportation and smart city. To mitigate the impact of FDI attacks, we propose a resilient control strategy that uses exponential ECBF and an observer-based auxiliary signal approach. This framework mitigates the impact of corrupted data, maintains system stability, and avoids collisions, even in the presence of cyberattacks. Practitioners in CPS design and deployment can benefit from this approach by integrating it into existing systems to improve safety and security. The proposed method is validated through both simulations and physical experiments, demonstrating its practicality for real-world applications.

This study examines the output tracking control of discrete-time networked semi-Markov jump systems (SMJSs) under cyber-attacks in the framework of finite-time control methodology. Different from most networked systems that employ single-channel communication, this work considers the case of multi-channel communication in the controller-to-actuator networks. Aimed at better reflecting the practical situation, a type of hybrid attacks is taken into consideration, which is a mixture of denial-of-service attacks and false data injection attacks. Subsequently, the dynamic characteristics of hybrid attacks among multiple channels are modeled by two stochastic processes. The goal is to design a state feedback controller such that the resulting closed-loop system is not only finite-time boundedness with dissipative performance but also has robustness against hybrid attacks. Finally, the effectiveness of the proposed novel controller design method is verified by an illustrative example.

The necessary and/or sufficient conditions for a Takagi-Sugeno-Kang Fuzzy Inference System (TSK-FIS) to be monotone has been a key research direction in the last two decades. In this article, we first define fuzzy membership functions (FMFs) with single and continuous support; and consider TSK-FIS with a grid partition strategy for computing its firing strengths with product T-norm (here after denoted as TSK-FIS-product). We also define a more general joint necessary condition, whereby each constituent itself is a necessary condition for the TSK-FIS-product model. The first necessary condition indicates that the normalized firing strength must not be indeterminate (i.e., 0/0), i.e., susceptible to the tomato classification problem. The second necessary condition indicates that all restricted consequents of fuzzy if-then rules must be defined. Based on the principle of the ordered weighted averaging (OWA) operator as well as the concept of increasing orness in OWA and hyperboxes, a general joint sufficient condition for a TSK-FIS-product model to be monotone is derived. Three case studies of the developed methods for undertaking failure mode and effect analysis (FMEA) and image processing tasks are presented. The results are compared, analyzed, and discussed, demonstrating the usefulness of our developed methods.

In this paper, we present Flood-Eye, an advanced framework for segmenting images intended for precise flood area estimation and segmentation. With its dual-pathway encoder, Flood-Eye is able to record contextual information at low resolution in addition to high-resolution details. To successfully integrate these features and improve the segmentation’s consistency, our model makes use of a novel cross-attention fusion technique. Furthermore, adaptive refinement is provided via a dynamic convolutional layer that adjusts responses to different flood patterns. A reliable component for estimating flooded areas is also included in the model, providing a quantitative analysis of the area of the flooded zone. With a mean intersection over union (mIoU) of 83.26% on the FloodNet dataset, Flood-Eye outperforms other state-of-the-art models in terms of performance and offers comprehensive and useful insights for efficient flood management and disaster response. This URL provides segmentation prediction and area estimation, which is drone footage of flooding in South Florida on June 14, 2024, and it clearly demonstrates the flooding’s impact and the need for advanced flood detection models like Flood-Eye. https://youtu.be/zltuXwrGUZM.

This article presents a novel learning-based collaborative control framework to ensure communication security and formation safety of nonlinear multiagent systems (MASs) subject to denial-of-service (DoS) attacks, model uncertainties, and barriers in environments. The framework has a distributed and decoupled design at the cyber-layer and the physical layer. A resilient control Lyapunov function-quadratic programming (RCLF-QP)-based observer is first proposed to achieve secure reference state estimation under DoS attacks at the cyber-layer. Based on deep reinforcement learning (RL) and control barrier function (CBF), a safety-critical formation controller is designed at the physical layer to ensure safe collaborations between uncertain agents in dynamic environments. The framework is applied to autonomous vehicles for area scanning formations with barriers in environments. The comparative experimental results demonstrate that the proposed framework can effectively improve the resilience and robustness of the system.

This work reports design problem of the observer-based integral sliding mode control for switched time-delay systems via asynchronous switching and dissipative condition. The main purpose of observer is to estimate the system’s entire state information based on measurement outputs, and further the observer states are used to create the controller. First, an improved free-weighting matrix inequality is introduced for concerning switched systems (SSs) to reduce conservatism of the integral terms. Then, a novel asymmetric Lyapunov–Krasovskii functional (LKF) is constructed for analysing the exponential stability of SSs with time-delay. With the help of improved integral inequality and novel LKF, a new set of sufficient conditions is developed in the form of linear matrix inequalities (LMIs). The developed criteria based on the sliding mode controller with observer scheme ensure that the proposed systems is exponentially stable with dissipative performance. Finally, numerical simulations are given to illustrate the usefulness and benefit of the proposed methods.

This paper explores the application of driving simulator across multiple sectors, highlighting the challenges associated with refining motion cueing algorithms (MCA) through model predictive control (MPC). Through these platforms, drivers can simulate the sensation of motion. The implementation of MPC-based MCA, while advantageous for its precision in controlling motion simulations, encounters significant hurdles such as the requirement for highly accurate system models and the extensive parameter tuning needed for each specific control scenario. These issues create a critical gap in achieving optimal simulation fidelity and efficiency with lower computational time, necessitating a novel approach to improve the MCA domain. Addressing these challenges, the study pioneers the use of Deep QNetwork (DQN), a reinforcement learning (RL) technique, to optimise the horizons of MPC within the MCA domain. This innovation is significant as it introduces, for the first time, a method to dynamically adjust MPC-based MCA horizons using DQN, which learns through continuous interaction with the simulation environment. This approach is set to overcome the limitations of traditional meta-heuristic optimisation methods, such as the Grasshopper Optimisation Algorithms (GOA) and Butterfly Optimisation Algorithms (BOA), by offering a more flexible and adaptable solution. The overarching goal of this research is to minimise the system's cost function by maximising a reward function that encompasses key performance metrics such as specific force sensation, angular velocity, linear displacement, linear velocity, and angular displacement. By integrating DQN into the MPC-based MCA environment, this study demonstrates a faster computational running time and improves the precision and efficiency of the simulations. This innovative approach enhances the efficiency of the horizon determination process, showcasing promising implications for the MCA domain's advancement.

Leveraging the concept of Failure Mode and Effect Analysis (FMEA), we propose a simple and systematic approach, namely Transmission Cause and Effect Analysis (TCEA), to achieve a defined goal of reducing transmission risk through effective preventive and control actions in real-world environments. Specifically, the transmission risk of an infectious disease (e.g., COVID-19) is perceived as a combination of the presence of a transmission agent (e.g., SARS-CoV-2 virus or its variants) and the requisite factors that lead to infection of humans and the associated aftermath of infection. TCEA adopts a causal map to represent all possible transmission risks via a brainstorming process. Next, appropriate preventive and control actions associated with each transmission risk are identified. Similar to FMEA, a Risk Priority Number model with Severity, Occurrence, and Detection ratings is adopted for analysis, prioritization, and decision-making. To demonstrate the usefulness of TCEA, a real-world case study on COVID-19 is conducted. The empirical results indicate that TCEA provide a simple, systematic and easy-to-implement approach to effectively analyze and manage transmission risks of COVID-19 in non-healthcare workplaces.

Cognitive load theory suggests that overloading of working memory may negatively affect the performance of human in cognitively demanding tasks. Evaluation of cognitive load is a difficult task; it is often assessed through feedback and evaluation from experts. Cognitive load classification based on Functional Near-InfraRed Spectroscopy (fNIRS) is now one of the key research areas in recent years, due to its resistance of artefacts, cost-effectiveness, and portability. To make fNIRS more practical in various applications, it is necessary to develop robust algorithms that can automatically classify fNIRS signals and less reliant on trained signals. Many of the analytical tools used in cognitive sciences have used Deep Learning (DL) modalities to uncover relevant information for mental workload classification. This review investigates the research questions on the design and overall effectiveness of DL as well as its key characteristics. We have identified 38 studies published between 2011 and 2022, that specifically proposed Machine Learning (ML) models for classifying cognitive load using data obtained from fNIRS devices. Those studies were analyzed based on type of feature selection methods, input, and DL model architectures. Most of the existing cognitive load studies are based on ML algorithms, which follow signal filtration and hand-crafted features. It is observed that hybrid DL architectures that integrate convolution and LSTM operators performed significantly better in comparison with other models. However, DL models especially hybrid models have not been extensively investigated for the classification of cognitive load captured by fNIRS devices. The current trends and challenges are highlighted to provide directions for the development of DL models pertaining to fNIRS research.

The recent breakthrough of deep learning based generative models has led to the escalated generation of photo-realistic synthetic videos with significant visual quality. Automated reliable detection of such forged videos requires the extraction of fine-grained discriminative spatial-temporal cues. To tackle such challenges, we propose weighted and evolving ensemble models comprising 3D Convolutional Neural Networks (CNNs) and CNN-Recurrent Neural Networks (RNNs) with Particle Swarm Optimization (PSO) based network topology and hyper-parameter optimization for video authenticity classification. A new PSO algorithm is proposed, which embeds Muller's method and fixed-point iteration based leader enhancement, reinforcement learning-based optimal search action selection, a petal spiral simulated search mechanism, and cross-breed elite signal generation based on adaptive geometric surfaces. The PSO variant optimizes the RNN topologies in CNN-RNN, as well as key learning configurations of 3D CNNs, with the attempt to extract effective discriminative spatial-temporal cues. Both weighted and evolving ensemble strategies are used for ensemble formulation with aforementioned optimized networks as base classifiers. In particular, the proposed PSO algorithm is used to identify optimal subsets of optimized base networks for dynamic ensemble generation to balance between ensemble complexity and performance. Evaluated using several well-known synthetic video datasets, our approach outperforms existing studies and various ensemble models devised by other search methods with statistical significance for video authenticity classification. The proposed PSO model also illustrates statistical superiority over a number of search methods for solving optimization problems pertaining to a variety of artificial landscapes with diverse geometrical layouts.

Abstract As complexity and capabilities of Artificial Intelligence technologies increase, so does its potential for misuse. Deepfake videos are an example. They are created with generative models which produce media that replicates the voices and faces of real people. Deepfake videos may be entertaining, but they may also put privacy and security at risk. A criminal may forge a video of a politician or another notable person in order to affect public opinions or deceive others. Approaches for detecting and protecting against these types of forgery must evolve as well as the methods of generation to ensure that proper information is supplied and to mitigate the risks associated with the fast evolution of deepfakes. This research exploits the effectiveness of deepfake detection algorithms with the application of a Particle Swarm Optimization (PSO) variant for hyperparameter selection. Since Convolutional Neural Networks excel in recognizing objects and patterns in visual data while Recurrent Neural Networks are proficient at handling sequential data, in this research, we propose a hybrid EfficientNet-Gated Recurrent Unit (GRU) network as well as EfficientNet-B0-based transfer learning for video forgery classification. A new PSO algorithm is proposed for hyperparameter search, which incorporates composite leaders and reinforcement learning-based search strategy allocation to mitigate premature convergence. To assess whether an image or a video is manipulated, both models are trained on datasets containing deepfake and genuine photographs and videos. The empirical results indicate that the proposed PSO-based EfficientNet-GRU and EfficientNet-B0 networks outperform the counterparts with manual and optimal learning configurations yielded by other search methods for several deepfake datasets.

Cervical cancer (CC) is affecting women predominantly, and early diagnosis could cure this cancer. This study aims to design and develop an effective deep learning-based classification model to detect early CC stages using clinical data. The proposed method is a combination of an unsupervised deep learning and a supervised neural network, i.e. sparse stacked autoencoder (SSAE) and fuzzy adaptive resonance theory MAP (FAM), respectively, and is denoted as SSAE-FAM. Specifically, SSAE is applied to tackle the data sparsity problem. It extracts the representative features from a data set through feature transformation. The transformed features are then classified by FAM. In this study, a CC data set obtained from the University of California Irvine (UCI) machine learning repository is utilised for evaluation. Owing to missing data in the original CC data set, two data sets are generated from the original CC data samples using two data preprocessing techniques. Both generated CC data sets with four target classes (i.e. Schiller, Cytology, Biopsy, and Hinselmann) are evaluated as four independent binary-class problems. We improve the classification performance of FAM by mitigating the data sparsity problem. Based on a series of experimental studies, SSAE-FAM outperforms other state-of-art methods by achieving 99.47%, 99.34%, 99.48%, and 99.81% mean accuracy rates, respectively, with the first CC data set, and 99.74%, 99.86%, 99.77%, and 99.80% mean accuracy rates, respectively, with the second CC data set. The results positively indicate the usefulness of SSAE-FAM for early CC diagnosis.

This article investigates the finite-time control problem of the switched affine systems via an event-triggered strategy. It is well known that the existence of affine terms brings great difficulties in analysis of the finite-time property of such systems. Furthermore, the design of the globally feasible event-triggered mechanism (ETM) under a finite-time control framework is challenging. Thus, a two-step hybrid control scheme is proposed in this article. The first step focuses on the event-triggered finite-time control for practical stability, while the second step aims to achieve finite-time stabilization. Particularly, in step one, by constructing the intersection between the affine term's threshold and feasible state region of the established ETM, it is verified that the Zeno behavior can be excluded. Thereafter, an affine state-dependent switching law and sufficient conditions are provided for achieving practical stability. Meanwhile, an estimation for the practical settling time to enter the bounded set is provided. In step two, the criteria for finite-time stabilization of the considered systems are further presented, and an overall settling-time upper bound is derived. Finally, a numerical example is illustrated to demonstrate the effectiveness of our proposed method.

This article studies the stability control problem of uncertain nonlinear systems with unknown multiple classes of actuator faults. We investigate not only traditional actuator gain and bias faults but also a class of novel actuator faults, i.e.,input power faults, which are induced by the change of system input powers. By combining fuzzy logic systems, adaptive control, and other control technologies and methods, we develop an adaptive fault-tolerant fuzzy control scheme that guarantees that the system under scrutiny is asymptotically stable and the system states asymptotically converge to an adjustable small neighborhood of the origin. Compared with the existing results, input power faults, as a class of novel actuator faults, are first proposed and investigated in this article, and the corresponding fault-tolerant control scheme is studied. Furthermore, even if the system is subject to unknown input power faults, the actuator faults, especially bias faults, can be compensated properly. Based on a mechanical system, the simulation results indicate the effectiveness of our proposed fault-tolerant control method.

This article aims to explore the dynamics involved in the intricate realm of chaotic synchronization in the Hindmarsh–Rose (H–R) neuronal model, which is known for its resemblance to the brain's information-processing components. Distinct from the existing studies related to the H–R neuronal model, this research focuses on addressing nonlinearities of membrane potential through a Takagi–Sugeno (T–S) fuzzy approach. A sampled-data-based controller scheme that can resolve the stabilization issues inherent to the H–R neuronal model is proposed. Compared with many existing control schemes, sampled-data control has several advantages, which include easy digital implementation and robustness against transmission delays. This research utilizes the zero-order holder technique to handle discrete-time control actions in a continuous-time T–S fuzzy model. Furthermore, synchronization analysis pertaining to the T–S fuzzy-based H–R model with user-designed control inputs is conducted to understand and overcome the associated spiking, chaotic, and bursting behaviors. Closed-loop dynamics of the resulting model with and without control input, namely, the error model, are analyzed by employing the Lyapunov stability theory and integral inequalities. Specifically, a suitable Lyapunov Krasovskii functional is formulated and solved using the linear matrix inequalities technique to guarantee the global asymptotically stability of the error model. The developed theoretical framework is validated using numerical simulations, and the corresponding outcomes are graphically illustrated and discussed.

The motion cueing algorithm (MCA) enables lifelike motion in simulators resembling real driving. Regenerated motions must adhere to workspace constraints. Vehicle motion signals (linear acceleration, angular velocity) are generated in a simulated vehicle environment utilised in MCA for motion cues. These signals are categorised into levels (slow, medium, fast) based on frequency and amplitude. The commonly used MCA, the classical washout filter, is typically fine-tuned using worst-case (fast-driving) scenarios to meet the simulator’s requirements across various situations. However, this approach reduces the MCA’s effectiveness in handling slower driving scenarios, resulting in conservatism in platform workspace usage for slow and medium driving. Consequently, a noticeable motion sensation error arises between real vehicle drivers and motion simulator users. To rectify this issue, a novel neural network-based MCA is developed in this study. Three distinct classical washout filters are meticulously tuned to cater to slow, medium, and fast driving scenarios. These filters generate precise motion cues for simulator users at corresponding levels of driving scenarios. The neural network-based MCA is constructed using the synthesised signals from these classical washout filters. This proposed method is thoroughly validated through the utilisation of MATLAB software. In direct comparison with the standard classical washout filter, the proposed MCA significantly reduces the motion sensation error, enriches motion fidelity, and optimises the utilisation of the simulator’s workspace.

Semantic segmentation is vital for many emerging surveillance applications, but current models cannot be relied upon to meet the required tolerance, particularly in complex tasks that involve multiple classes and varied environments. To improve performance, we propose a novel algorithm, neural inference search (NIS), for hyperparameter optimization pertaining to established deep learning segmentation models in conjunction with a new multiloss function. It incorporates three novel search behaviors, i.e., Maximized Standard Deviation Velocity Prediction, Local Best Velocity Prediction, and n -dimensional Whirlpool Search. The first two behaviors are exploratory, leveraging long short-term memory (LSTM)-convolutional neural network (CNN)-based velocity predictions, while the third employs n -dimensional matrix rotation for local exploitation. A scheduling mechanism is also introduced in NIS to manage the contributions of these three novel search behaviors in stages. NIS optimizes learning and multiloss parameters simultaneously. Compared with state-of-the-art segmentation methods and those optimized with other well-known search algorithms, NIS-optimized models show significant improvements across multiple performance metrics on five segmentation datasets. NIS also reliably yields better solutions as compared with a variety of search methods for solving numerical benchmark functions.

The problem of resilient optimal bipartite tracking control for heterogeneous multi-agent systems with multiple targets under denial-of-service (DoS) attacks is investigated in this paper. A bipartite tracking mechanism is devised in which the agents track the targets under bipartite consensus control, which becomes non-autonomous. This is owing to DoS attacks, as the agents cannot obtain real-time information on the tracked targets and neighboring agents. Consequently, A target observer with a storage module has been developed for the efficient estimation of agent states and the storage of observed information as historical data. By recalling the historical data of the observed state, a new type of distributed resilient optimal controller is formulated, which can achieve the control objective in the case of communication blockage, while minimizing the performance index function of the system. Numerical simulations are performed to verify the proposed secure control design.

Neural networks (NNs) play a significant role in the machine learning and deep learning domains that include pattern recognition, computer-vision and so on. However, understanding the theoretical properties of neural networks will helps to deliver the user-desired performance in such practical applications. In the literature, the fundamental analysis of a NN, such as stability analysis, parameter sensitivity analysis can be performed by modeling the neuronal activities as differential equations. Through differential equations, the rate at which information is transmitted can be experimented along with various significant factors, such as time-delays during data transmission, switching parameters with respect to time, random disturbances caused by interruption of data blocks. The present study focuses on fundamental analysis of neuronal activities through differential model. Besides, the factors, such as time-delays, exogenous disturbances, and Markovian-jumping parameter (MJP) that has an ability to degrade the stable performance of the neuronal model is incorporated in the model. Distinct to the previous studies in stochastic neural networks, the study address the synchronization problem of stochastic neural networks (SNNs) with fractional-derivative of Brownian motion and event-triggered control scheme. Theoretically, due to nonlinearties, the Lyapunov stability theory is employed to derive the sufficient stability conditions that ensure the stable performance of SNNs. In this regard, looped-Lyapunov functional candidate is considered and corresponding linear matrix inequalitys (LMIs) are derived. Technically, a model of two neurons, three neurons, and four neurons are considered with the given factors to validate the proposed theoretical conditions and controller performance and their results are picturised.

Precise and reliable measurement of wind power uncertainty plays a significant role in the economic operation and real-time control of the smart grid. In this paper, a novel spiking neural network (SNN) architecture is proposed for solving regression tasks, and a multi-objective gradient descent (MOGD) algorithm is employed to generate high-quality wind power prediction intervals (PIs). SNNs improve upon conventional artificial neural networks (ANNs) by encoding interneuron communication into temporally-distributed spikes, which reduce memory access frequency and data communication, and therefore, the computational power requirements of deep learning workloads. This becomes exceedingly important for continual data analysis in remote geographic regions which often lack reliable cloud access and power supply, where many wind power farms are stationed. Given that neuron spikes are all stereotypically treated to be identical, they are a natural fit for tasks that may conflict in a common network architecture, such as multimodal data or where multiple, potentially competing, objectives are being optimized for. This paper proposes an SNN architecture that achieves comparable performance with its ANN counterpart on a complex regression task, i.e., wind power interval prediction. The resulting multi-objective SNN demonstrates superior performance as compared with those from state-of-art ANNs in wind power interval prediction.

Autonomous vehicles, commonly known as self-driving cars, are rapidly gaining popularity due to their numerous advantages, such as reducing traffic, pollution, and emissions while increasing safety, convenience, and transportation connectivity. In order to accurately track the motion signal, these vehicles are now utilising advanced control techniques, such as model predictive control (MPC). However, the efficiency of MPCs heavily relies on properly tuning their weights. The primary function of the MPC is to recalculate the optimal values for the vehicle control commands, such as desired speed, steering angle, etc., while considering the dynamic model of the autonomous vehicle. The existing linear MPC models cannot reach higher efficiency because of using fixed weights without considering the error. This paper introduces a novel approach for developing an MPC model with a time-varying weights algorithm for autonomous vehicles. The study aims to minimise motion tracking errors such as lateral position and yaw angle errors. Relevant MPC weights are calculated online using fuzzy logic-based units considering the lateral position and yaw angle errors. The proposed linear time-varying MPC was designed and developed using MATLAB software, resulting in improved motion tracking performance with 31.62% and 20.89% reduction of the root means square error of lateral position and yaw angle.

Abstract Reinforcement learning (RL) has emerged as an effective approach for building an intelligent system, which involves multiple self-operated agents to collectively accomplish a designated task. More importantly, there has been a renewed focus on RL since the introduction of deep learning that essentially makes RL feasible to operate in high-dimensional environments. However, there are many diversified research directions in the current literature, such as multi-agent and multi-objective learning, and human-machine interactions. Therefore, in this paper, we propose a comprehensive software architecture that not only plays a vital role in designing a connect-the-dots deep RL architecture but also provides a guideline to develop a realistic RL application in a short time span. By inheriting the proposed architecture, software managers can foresee any challenges when designing a deep RL-based system. As a result, they can expedite the design process and actively control every stage of software development, which is especially critical in agile development environments. For this reason, we design a deep RL-based framework that strictly ensures flexibility, robustness, and scalability. To enforce generalization, the proposed architecture also does not depend on a specific RL algorithm, a network configuration, the number of agents, or the type of agents.

Recently, speech emotion recognition (SER) has become an active research area in speech processing, particularly with the advent of deep learning (DL). Numerous DL-based methods have been proposed for SER. However, most of the existing DL-based models are complex and require a large amounts of data to achieve a good performance. In this study, a new framework of deep attention-based dilated convolutional-recurrent neural networks coupled with a hybrid data augmentation method was proposed for addressing SER tasks. The hybrid data augmentation method constitutes an upsampling technique for generating more speech data samples based on the traditional and generative adversarial network approaches. By leveraging both convolutional and recurrent neural networks in a dilated form along with an attention mechanism, the proposed DL framework can extract high-level representations from three-dimensional log Mel spectrogram features. Dilated convolutional neural networks acquire larger receptive fields, whereas dilated recurrent neural networks overcome complex dependencies as well as the vanishing and exploding gradient issues. Furthermore, the loss functions are reconfigured by combining the SoftMax loss and the center-based losses to classify various emotional states. The proposed framework was implemented using the Python programming language and the TensorFlow deep learning library. To validate the proposed framework, the EmoDB and ERC benchmark datasets, which are imbalanced and/or small datasets, were employed. The experimental results indicate that the proposed framework outperforms other related state-of-the-art methods, yielding the highest unweighted recall rates of 88.03 ± 1.39 (%) and 66.56 ± 0.67 (%) for the EmoDB and ERC datasets, respectively.

In this research, we propose a variant of the Bare-Bones Particle Swarm Optimization (BBPSO) algorithm for hyper-parameter selection and deep architecture generation for image, audio and video classification tasks. Since the search process of the original BBPSO model is guided by a single leader and the particles’ personal best experiences, there is a lack of interactions pertaining to the neighbouring elite solutions. To overcome this limitation, we propose a versatile search process for a modified BBPSO model that incorporates a number of effective components and operations. These include the neighbouring and global best signals, search actions with Cauchy/Levy scale factors, sub-dimension operations guided by the local and global elite solutions, and a Levy-driven local search mechanism. Moreover, root-finding algorithms are employed which use informative mathematical principles to estimate new root offspring for leader/particle enhancement. A reinforcement learning algorithm is subsequently used to identify the optimal sequential deployment of these numerical analysis methods to increase robustness. Several medical imaging data sets, i.e., ISIC 2017, PH2 and Dermofit skin lesion databases, the ALL-IDB2 microscopic blood image data set, the MURA musculoskeletal radiographic database, the CK + facial expression data set, as well as the Coswara respiratory audio data set and UCF101 video action data set, are employed for evaluation. The proposed BBPSO-optimized Convolutional Neural Network (CNN), bidirectional Long Short-Term Memory (BiLSTM) with attention mechanism, and CNN-BiLSTM models outperform those devised by other PSO and BBPSO variants, as well as state-of-the-art existing studies, significantly, for image, audio respiratory abnormality and realistic video action recognition.

Automatic segmentation of salient objects in real-world images has gained increasing interests owing to its popularity in diverse real-world applications, such as autonomous driving, medical diagnosis, aviation security, and underwater surveillance. In this research, we propose Firefly Algorithm (FA)-enhanced evolving ensemble deep networks for semantic segmentation and visual saliency prediction. An improved FA model is proposed to optimize network hyper-parameters. Specifically, it employs mutation operators and a neighbouring search strategy with granular search steps to establish search intensification. It also emphasizes search diversification by adopting multiple dynamic hybrid leaders and diverse adaptive sine and cosine search trajectories in full and randomly selected sub-dimensions to overcome stagnation. Because of its competent segmentation performance, DeepLabV3+ is fine-tuned using transfer learning with FA-based hyper-parameter identification. We optimize the learning rate, momentum and weight decay of the transfer learning network. A number of optimized DeepLabV3+ networks with distinguishing learning configurations are yielded. An ensemble model is subsequently constructed by incorporating three optimized base networks to further strengthen segmentation performance. Evaluated using diverse challenging semantic segmentation and saliency prediction tasks using underwater and medical image data sets, our evolving ensemble deep network illustrates significant superiority over other state-of-the-art deep networks and existing studies. The proposed FA model also outperforms other search methods in solving diverse mathematical landscapes with statistical significance.

Generalized zero-shot learning (GZSL) aims to train a model for classifying data samples under the condition that some output classes are unknown during supervised learning. To address this challenging task, GZSL leverages semantic information of the seen (source) and unseen (target) classes to bridge the gap between both seen and unseen classes. Since its introduction, many GZSL models have been formulated. In this review paper, we present a comprehensive review on GZSL. First, we provide an overview of GZSL including the problems and challenges. Then, we introduce a hierarchical categorization for the GZSL methods and discuss the representative methods in each category. In addition, we discuss the available benchmark data sets and applications of GZSL, along with a discussion on the research gaps and directions for future investigations.

Abstract This paper proposes an enhancement to the Harris’ Hawks Optimisation (HHO) algorithm. Firstly, an enhanced HHO (EHHO) model is developed to solve single-objective optimisation problems (SOPs). EHHO is then further extended to a multi-objective EHHO (MO-EHHO) model to solve multi-objective optimisation problems (MOPs). In EHHO, a nonlinear exploration factor is formulated to replace the original linear exploration method, which improves the exploration capability and facilitate the transition from exploration to exploitation. In addition, the Differential Evolution (DE) scheme is incorporated into EHHO to generate diverse individuals. To replace the DE mutation factor, a chaos strategy that increases randomness to cover wider search areas is adopted. The non-dominated sorting method with the crowding distance is leveraged in MO-EHHO, while a mutation mechanism is employed to increase the diversity of individuals in the external archive for addressing MOPs. Benchmark SOPs and MOPs are used to evaluate EHHO and MO-EHHO models, respectively. The sign test is employed to ascertain the performance of EHHO and MO-EHHO from the statistical perspective. Based on the average ranking method, EHHO and MO-EHHO indicate their efficacy in tackling SOPs and MOPs, as compared with those from the original HHO algorithm, its variants, and many other established evolutionary algorithms.

Nonlinear model predictive control has been used in motion cueing algorithms recently to consider the nonlinear dynamics model of the system. The entire motion cueing algorithm indexes, including the physical and dynamical constraints of the actuators and physical constraints of passive joints, can be controlled with precision using nonlinear model predictive control. However, several weighting parameters in the nonlinear model predictive control-based motion cueing algorithm (including driving sensation, motion description of the actuators, and passive joints) require proper and laborious tuning to attain an optimal design structure. In this work, the optimal weighting parameters of a nonlinear predictive control-based motion cueing algorithm model are calculated using cascade optimisation and human interaction. A cascade optimisation method consisting of a particle swarm optimisation and genetic algorithm is designed to identify the best weighting parameters compared to those from one optimiser. In addition, the human decision-making units are added to the two-level cascade optimiser to determine the best solution from a Pareto front. The proposed cascade optimiser decreases the run-time with better extraction of the optimal weighting parameters to increase the motion fidelity compared to a single optimiser. It should be noted that the proposed methodology is applied along longitudinal channel. While the same methodology can be applied along lateral, heave and yaw channels for further evaluation of the proposed method. The proposed model is simulated utilising the MATLAB software and the results prove the efficiency of the newly proposed model compared to those from the previous single optimiser in reproducing more accurate motion signals with better usage of the driving motion platform workspace.

Advances in utilisng motion simulators for skill training and related applications have yielded numerous benefits, such as safety, availability, and serviceability, environmentally friendly, and economically beneficial. To give simulator users a sense of realistic feeling of driving, an accurate motion cueing algorithm (MCA) is essential, in order to respect the simulator platform limitation and avoid motion sickness. The use of Model Predictive Control (MPC) in MCA designs leads to respecting the constraints and considering the future dynamic behaviors of the simulator. However, the tuning process of the MPC prediction horizon and control horizon still need to be improved. These horizons are normally selected manually by the designer. Previous studies on meta-heuristic algorithms produce a large prediction horizon with a heavy computational load or a small prediction horizon that sacrifices the stability and accuracy of the simulator system. In this study, the Grasshopper Optimization Algorithm (GOA) is adopted to yield optimal prediction and control horizons in MPC-based MCA models. The results are compared with those from the Butterfly Optimization Algorithm (BOA) and Genetic Algorithm (GA) in terms of sensation error and computation time. The GOA technique depicts the fastest process time to promptly detect proper MPC horizons. It does not affect the simulator's efficiency in utilising the workspace, as evidenced by the correlation coefficient and root mean square error between sensation from a real-world vehicle and the simulator.

Self-driving vehicles, also known as Autonomous Vehicles (AVs), are steadily becoming very popular due to their huge benefits. They can improve safety, convenience and transport interconnectivity as well as reduce congestion, pollution and emissions. The generation of the comfort motion signal for AVs passenger via the calculation of accurate motion cues with lower motion discomforts is important to promote the adoption of Avs in society. Model predictive control (MPC) is currently used in AVs for tracking the motion signal with good accuracy. However, the higher efficiency of MPC is directly related to the right setting of the weights. In addition, the tracking of time-varying longitudinal velocity is not possible without using linear time-varying (LTV) MPC. In this study, an LTV MPC system is designed and developed as a highly efficient motion tracking mechanism for AVs to reduce the motion tracking error and motion discomfort. In addition, biogeography-based optimisation (BBO) is employed to determine the optimal weights of the LTV MPC controller, which further reduces the motion tracking error and increases the motion comfort for users. The empirical study demonstrates that a BBO-tuned LTV MPC controller decreases the mean square error of motion tracking by 4.79% as compared with that of a manually-tuned version. Moreover, the mean square errors of the lateral deviation and relative yaw decrease by 91.22% and 19.14% as compared with those from a manually-tuned LTV MPC counterpart, respectively.

Motion signal should be generated via the AV control system targeting the maximum motion comfort for the users. Nonlinear model predictive control (MPC) is recently used in AVs to achieve this critical task. However, nonlinear MPC has lots of hyperparameters, including weights and MPC horizons, that should be tuned systematically to reach the system’s high efficiency. The energy usage and motion comfort have a direct relationship. The generation of high-fidelity motion cues for AV users leads to higher energy usage. Hence, there is a need for the use of a multi-objective optimisation technique to tune the weights wisely to satisfy the appropriate energy usage and motion comfort for the AV users. In this study, multi-objective NSGA-II is employed, for the first time, to tune the weights of a nonlinear MPC-based controller in AVs. The proposed method is designed and developed using MATLAB/SIMULINK software. The simulation results show minimum energy usage by generation of smooth motion signals, delivering maximum comfort to AV users.

The autonomous vehicle (AV) has been studied by many researchers recently because of its valuable points in transportation, aviation, military, smart city, and aerospace. The model predictive control (MPC) is employed to track the artificial intelligent regenerated motion signals with higher accuracy than other error-and model-based controllers as it can consider the constraints of the system in extracting the optimal solution. However, the accuracy and applicability of the MPC rely on the MPC horizons, including prediction and control horizons. The higher prediction horizons mean a higher computational load of the system, which reduces the real-time applicability of the system. On the other hand, a higher prediction horizon increases the system’s stability in facing abrupt motion signals. In addition, higher control horizons mean more dexterity in the system facing an unknown situation. On the other hand, a longer control horizon increases the computational load of the system exponentially. This study employs particle swarm optimisation (PSO) to extract the optimal MPC horizons considering the accuracy and computational load. The cost function is defined to increase the accuracy of the longitudinal time-varying velocity tracking, decrease the lateral deviation, decrease the relative yaw angle and decrease the computational load of the system. It should be noted that the lateral deviation and relative yaw angle are extracted using the vehicle four wheels dynamic model in order to evaluate the AVs’ passenger motion comfort. The proposed method is designed and developed under MATLAB/Simulink. The extracted optimal MPC horizon is compared with some other arrangements of the MPC horizons to prove the efficiency of the proposed method compared with the trial-and-error method.

The motion signals are generated for a simulator user based on the visual understanding of the environment using virtual reality. In this respect, a motion cueing algorithm (MCA) is employed to reproduce the motion signals based on the real driving motion scenarios. Advanced MCAs are required to predict precise driving motion scenarios. Nonetheless, investigations on effective methods for predicting the driving motion scenarios accurately are limited. Current state-of-the-art studies mainly focus on the averaged motion signals from several simulator users pertaining to a specific map or from feedforward neural network and non-linear autoregressive. The existing methods are unable to yield precise predictions of the driving scenarios. In this research, the echo state network and long short-term memory models are employed for the first time in MCA to forecast the driving motion signals. Our evaluation proves the efficiency of our proposed methods in comparison with existing methods.

Abstract Still image human action recognition (HAR) is a challenging problem owing to limited sources of information and large intra-class and small inter-class variations which requires highly discriminative features. Transfer learning offers the necessary capabilities in producing such features by preserving prior knowledge while learning new representations. However, optimally identifying dynamic numbers of re-trainable layers in the transfer learning process poses a challenge. In this study, we aim to automate the process of optimal configuration identification. Specifically, we propose a novel particle swarm optimisation (PSO) variant, denoted as EnvPSO, for optimal hyper-parameter selection in the transfer learning process with respect to HAR tasks with still images. It incorporates Gaussian fitness surface prediction and exponential search coefficients to overcome stagnation. It optimises the learning rate, batch size, and number of re-trained layers of a pre-trained convolutional neural network (CNN). To overcome bias of single optimised networks, an ensemble model with three optimised CNN streams is introduced. The first and second streams employ raw images and segmentation masks yielded by mask R-CNN as inputs, while the third stream fuses a pair of networks with raw image and saliency maps as inputs, respectively. The final prediction results are obtained by computing the average of class predictions from all three streams. By leveraging differences between learned representations within optimised streams, our ensemble model outperforms counterparts devised by PSO and other state-of-the-art methods for HAR. In addition, evaluated using diverse artificial landscape functions, EnvPSO performs better than other search methods with statistically significant difference in performance.

Summary The complex and large‐scale scientific workflow applications are effectively executes on the cloud. The performance of cloud computing highly depends on the task scheduling. Optimal workflow scheduling is still a challenge that needs to be addressed due to the conflicting objectives and increasing demand for quality of service. Task scheduling is an NP‐hard problem due to its complexity. The newly introduced methods for resolving the problem of task scheduling are facing challenges to take the benefits of all aspects of cloud computing. In this article, we study the joint optimization of cost and makespan of scheduling workflows in infrastructure as a service clouds and propose a new workflow scheduling scheme using deep learning. In this scheme, a deep‐Q learning‐based heterogeneous earliest‐finish‐time (DQ‐HEFT) algorithm is developed, which closely integrates the deep learning mechanism with the task scheduling heuristic HEFT. The workflowsim simulator is used for the experiment of the real‐world and synthetic workflows. The experiment results demonstrate the efficiency of our proposed approach compared with existing algorithms. This technique can achieve significantly better makespan and speed metrics with a remarkably higher volume of data and can run faster compared with the existing workflow scheduling algorithms in cloud computing environment.

In this paper, the resilient controller design and synthesis issues of networked control systems under denial-of-service (DoS) attacks are investigated via an adaptive event-triggered strategy. In a networked control system, DoS attacks have serious impacts on the security of communication, possibly causing degraded stability performance of the system. To remove threats from DoS attacks, an adaptive event-triggered communication mechanism is proposed, which also provides benefits in reducing the consumption of communication resources and relieving the pressure on network bandwidth. Since the system state information is usually not fully known, an observer-based controller is developed to stabilize the system and maintain a desired performance index despite the occurrence of stochastic attacks. Furthermore, a joint design method is proposed to obtain the controller gain, observer gain and event-triggered weight matrix. Finally, practical examples based on a four-tank system, an Internet-based three-tank system and an Internet-based test rig system are presented to illustrate the effectiveness of the proposed techniques to respond and eliminate the impact of DoS attacks.

Formulating a generalized monotone fuzzy rule interpolation (MFRI) model is difficult. A complete and monotone fuzzy rule-base is essential for devising a monotone zero-order Takagi–Sugeno–Kang (TSK) fuzzy inference system (FIS) model. However, such a complete and monotone fuzzy rule-base is not always available in practice. In this article, we develop an MFRI modeling scheme for generating a monotone zero-order TSK FIS from a monotone and incomplete fuzzy rule-base. In our proposal, a monotone-ordered fuzzy rule-base that consists of the available fuzzy rules from a monotone and incomplete fuzzy rule-base and those derived from the MFRI reasoning is formed. We outline three important properties that the MFRI's deduced fuzzy rules should satisfy to ensure a monotone-ordered fuzzy rule-base. A Lagrangian function for the MFRI scheme, together with its Karush–Kuhn–Tucker optimality conditions, is formulated and analyzed. The key idea is to impose constraints that guide the MFRI inference outcomes. An iterative MFRI algorithm that adopts an augmented Lagrangian function is devised. The proposed MFRI algorithm aims to achieve an${\boldsymbol{\varepsilon }}$-optimality condition and to produce an${\boldsymbol{\varepsilon }}$-optimal solution, which is geared for practical applications. We apply the MFRI algorithm to a failure mode and effect analysis case study and a tanker ship heading regulation problem. The results indicate the effectiveness of MFRI for generating monotone TSK FRI models in tackling practical problems.

This article addresses the event-triggered tracking control and filtering problem for Takagi–Sugeno fuzzy-approximation-based discrete-time nonlinear networked systems subject to the effect of denial-of-service attacks. First, the unreliability of the communication channel between the sensor and the actuator/filter is considered, and the packet loss caused by denial-of-service attacks is characterized by the binary Markov chain. Second, two novel resilient adaptive event-triggered mechanisms are proposed to resist the impact of denial-of-service attacks and discard unnecessary data packets, in which the dynamic threshold variable is designed to adjust the event-triggered condition adaptively. A problem caused by the event-triggered mechanism is that the controller or filter cannot receive the system mode signal during the trigger interval. To surmount this problem, this article designs an estimator to compensate for the unavailable system mode. Then, an adaptive event-triggered controller or filter related to the estimated mode is designed to track the desired signal. Under this framework, by constructing a membership-function-dependent Lyapunov function, the conservativeness of the stability criterion is relaxed. Finally, two examples are used to validate the applicability of the proposed approaches.

Learning to drive in urban areas is an open challenge for autonomous vehicles (AVs), as complex decision making requirements are needed in multi-task co-ordinations environments. In this paper, we propose a hybrid framework with a new perception model involving affordance learning to simplify the surrounding urban scenes for training an AV agent, along with a planned trajectory and the associated driving measurements. Our proposed solution encompasses two main aspects. Firstly, a supervised learning network is used to map the input sensory data into affordance predictions. The predicted affordances provide a low-dimensional representation of surrounding scenes of the AV in the form of key perception indicators, e.g., true or false with respect to a traffic light signal. Secondly, a deep deterministic policy gradient model that maps the perception information into a series of actions is devised. We evaluate the proposed solution using the CARLA driving simulator in an urban town and evaluate the performance in a new, unseen town under different weather conditions. The quantitative and qualitative results indicate that our proposed solution can generalize well to cope with different traffic solutions and environmental conditions. Our proposed solution also outperforms other baseline methods in a comparative study in handling various AV driving tasks with different levels of difficulty. In addition, the model trained with simulated scenes yields promising prediction results when testing on recorded video streams on real-world highway and suburb environments with varying traffic and weather conditions.

The motion cueing algorithm (MCA) is playing the most critical role in motion simulation platform (MSP) to reproduce the realistic motion sensation of the real car for the MSP’s users while taking into cogitation of the physical boundaries of the platform. Recently, the model predictive control (MPC) is employed for designing MCAs which led to the formation of MPC-based MCAs. The purpose of the MPC-based MCA is to recalculate the optimal values of the input signals with consideration of the MSP’s physical restrictions. All the current MPC-based MCAs have fix weights that can cause the conservative and inefficient utilization of the MSP’s workspace limitations as they have been tuned based on the worst-case scenarios to keep the platforms within their physical limitations. Then, the error of motion sensation between the real car and MSP users increases due to the conservative utilization of the MSP’s workspace boundaries. The main objective of this study is to provide more efficient workspace utilization to minimise the error of motion feeling between the real car and MSP users while respecting the workspace boundaries. A procedure according to the optimised fuzzy logic-based units is employed to calculate the appropriate MPC weights online while considering the sensed specific force error, sensed angular velocity error and the current motion status of the MSP including linear position, linear velocity and angular position of the cockpit. The proposed MPC-based MCA is designed and developed using MATLAB. The outcomes show a better motion feeling compared with the current MPC-based MCAs.

The motion cueing algorithm (MCA) is the main algorithm in motion simulators in charge of generating vehicle motions within the platform’s constraints. The classical washout filter is one of the popular types of MCA, which is used in air and land vehicle motion simulators. The fixed home position of the simulator platform is always cogitated in the MCA to washout the motion simulator after generating each motion. Unfortunately, considering the fixed home position reduces the efficient consumption of the workspace in the linear directions. The linear motion of the motion simulator is due to the production of the high-pass frequency part of the motion scenarios. Prepositioning is used to tackle this assumption by varying the home position rather than the fixed position. The linear motion limitations of the motion simulator can virtually be enlarged using the prepositioning method. The efficient regeneration of the high-pass motion cues using a new propositioning technique is the main goal of this study to increase the motion realism of the simulator and remove any false motion cues due to the platform limitations. The proposed model utilised the recurrent neural network (RNN) to estimate the motion scenario along the prediction horizon. The nonlinear model predictive control (MPC) uses the estimated motion signals to extract the best optimal off-centre position of the motion simulator platform. The newly developed prepositioning technique is developed in the simulation environment of MATLAB to validate the proposed technique in terms of efficiency and applicability. The outcomes prove the capability of the proposed technique against the recently developed prepositioning technique using fuzzy logic and RNN.

Learning to drive in highly crowded urban areas with multi-agent interaction is a challenging task. Most of the current methods rely on hand-crafting the policy required for the decision-making process. In contrast, reinforcement learning (RL) offers proper methodologies to automatically formulate the best policy without manual intervention. However, the baseline RL algorithms face a challenge in handling simulated driving situations (e.g., keep driving in the current lane center, and keep a safe distance from a leading vehicle) in high fidelity and complex urban areas. In this study, we propose an end-to-end autonomous driving system using a Deep Q-Network (DQN) and long-short-term memory (LSTM) with a new observation input to enable the RL agent to learn driving in complex environments i.e. CARLA simulator. The observation input comprises a tuple of RGB (Red, Green, Blue) image data from the forward-facing camera, vehicle speed, and vehicle angle from the road center. The output comprises the steering, brake, and acceleration commands. Our proposed model, in conjunction with formulating the proper reward function, results in a rapid convergence with a better performance and safer driving behaviors from an autonomous driving agent. The results indicate that the LSTM-DQN model enables the autonomous agent in controlling the vehicle in a lane while keeping a safe distance from the leading vehicle in varied unseen road conditions.

The purpose of the motion cueing algorithm is to reproduce the motion sensation for the drivers considering the physical limitations of this platform. Newly, the model predictive control-based methods have been used in motion cueing algorithms. This control respects the constraints and considers the future dynamics of the model for finding the optimum solution to the problem. However, the tuning of the weights for model predictive control is incredibly challenging to reduce the motion sensation errors. In this paper, a whale optimisation algorithm is used to gain the optimised weights of the model predictive control. The weights are optimized to reduce the cost function which is defined based on the motion inputs, input rates, and outputs. The recalculated weights via the whale optimisation algorithm should consider the limitations of the applications such as maximum tolerated error of motion sensation via the motion platform user, maximum linear and angular displacements, and maximum linear velocity. The proposed method is simulated seven times to demonstrate the accuracy and repeatability of the algorithm. The results show that the whale optimisation algorithm reaches the best solution quickly with minimised motion sensation error compared with the genetic algorithm.

The usage of autonomous vehicles in the transportation sector can achieve the objective of a safe environment. To increase riding comfort in an autonomous vehicle, one main challenge is to implement motion scenarios according to the passenger’s driving behaviours. This leads to customization of the driving style of an autonomous vehicle according to the preference of its passenger. The main disadvantage of the current autonomous vehicles is the regeneration of driving motion signals without taking into consideration the comfort/discomfort of the passengers according to their driving behaviours and preferred driving styles such as acceleration/deceleration rate and steering styles. In this paper, a nonlinear autoregressive network model is developed and trained based on the generated motion scenarios of the passenger and the position of the autonomous vehicle, in order to predict and replicate the motion signals based on the passenger’s driving behaviours. The MATLAB toolbox is used to train the network and forecast the motion signals. The results show the usefulness of the proposed method in terms of a higher shape similarity level and a lower mean square error rate between the actual and forecasted motion signals. These regenerated motion signals can increase the riding comfort of autonomous vehicle’s passengers as it is able to imitate the behaviour of the passengers.

The motion simulator platform can be used in many sectors, including transportation, aviation, and education. The motion cueing algorithm (MCA) is the main component of the motion simulator with the responsibility of motion cues regeneration while respecting the motion simulator’s joint limitations. Recently, a model predictive control (MPC) method has been introduced in the MCA, which is able to extract an optimum input signal within the model constraints. The washout speed of the end-effector using the existing MPC-based MCA model is not considered because the integral of linear displacement of the platform is omitted as an output. As a result, the motion simulator platform returns to the neutral position without the consideration of the motion behavior. In this study, the integral of the end-effector linear displacement is consider inside the MPC-based MCA model to select the best washout speed of the end-effector. Moreover, a grey wolf optimizer is utilized to identify the best MPC weighting indexes, in order to increase the model efficiency for both existing and proposed models. The proposed method outperforms the MPC-based MCA model in producing better regeneration of the motion cues. It yields a higher correlation coefficient and a lower root means square error between the motion sensation signal pertaining to the real vehicle and motion simulator platform users.

System-of-Systems design has focused on integrated service offerings. The underlying expectation is that services have a reliance on each other to achieve a greater level of service and outcomes. Within this context, Cloud Computing provides a service consideration that extends system-of-systems design and thinking. Integration with cross provider services, operational and architectural design, systems operations, trust, and cost analysis must all be considered when addressing complex, cloud-based application design. This paper opens a discussion on these key attributes and provides a framework for future research whilst introducing key operational requirements to enable successful implementation of cross hyperscaler system-of-systems architectural models.

Despite the popularity and practical importance of the fuzzy inference system (FIS), the use of an FIS model as ann-ary aggregation function, which is characterized by both the monotonicity and boundary properties, is yet to be established. This is because research on ensuring that FIS models satisfy the monotonicity property, i.e., monotone FIS, is relatively new, not to mention the additional requirement of satisfying the boundary property. The aim of this article, therefore, is to establish the parametric conditions for the Takagi–Sugeno–Kang (TSK) FIS model to operate as ann-ary aggregation function (hereafter denoted asn-TSK-FIS) via the specifications of fuzzy membership functions and fuzzy rules. An absorption property with fuzzy rules interpretation is outlined, and the use ofn-TSK-FIS as a uninorm is explained. Exploiting the established parametric conditions, a framework for which ann-TSK-FIS model can be constructed from data samples is formulated and analyzed, along with a number of remarks. Synthetic data sets and a benchmark example on education assessment are presented and discussed. To be best of the authors’ knowledge, this article serves as the first use of the TSK-FIS model as ann-ary aggregation function.

The fuzzy min-max (FMM) neural network stands as a useful model for solving pattern classification problems. FMM has many important features, such as online learning and one-pass learning. It, however, has certain limitations, especially in its learning algorithm, which consists of the expansion, overlap test, and contraction procedures. This article proposes a refined fuzzy min-max (RFMM) neural network with new procedures for tackling the key limitations of FMM. RFMM has a number of contributions. First, a new expansion procedure for overcoming the problems of overlap leniency and irregularity of hyperbox expansion is introduced. It avoids the overlap cases between hyperboxes from different classes, reducing the number of overlap cases to one (containment case). Second, a new formula that simplifies the original rules in the overlap test is proposed. It has two important features: (i) identifying the overlap leniency problem during the expansion procedure; (ii) activating the contraction procedure to eliminate the containment case. Third, a new contraction procedure for overcoming the data distortion problem and providing more accurate decision boundaries for the contracted hyperboxes is proposed. Fourth, a new prediction strategy that combines both membership function and distance measure to prevent any possible random decision-making during the test stage is proposed. The performance of RFMM is evaluated with the UCI benchmark datasets. The results demonstrate the effectiveness of the proposed modifications in making RFMM a useful model for solving pattern classification problems, as compared with other existing FMM and non-FMM classifiers.

Driving motion simulators are widely used for their reliable, safe and cost-effective abilities to replicate real vehicle driving experience for simulator drivers in virtual environment. As all motion simulators have physical limitations, Motion Cueing Algorithm (MCA) is the most necessary algorithm for transformation of the real vehicle's linear and rotational motions to motion platform aiming to regenerate realistic driving sensation. Model Predictive Control (MPC)-based MCA has recently become one of the most popular MCAs. Scaling and limiting is an important unit of MPC-based MCA to reduce the amplitude of motion signal uniformly aiming to improve the realism of produced motion within the physical limitations of workspace. The current implementations of MPC use a basic form of scaling. In this paper, a novel MPC-based MCA is developed using an optimised nonlinear scaling unit and Genetic Algorithm (GA). The goal is to reproduce accurate motion sensation for the motion simulator drivers as close as possible to real vehicle within the platform's physical constraints. This is achieved via a polynomial scaling unit which is optimized by GA. The aim is to overcome the disadvantages associated with the tuning based on trial-and-error for MPC-based MCA scaling unit which is the main cause of inefficient platform workspace usage and motion sensation error between real vehicle driver and motion simulator driver. The proposed optimization-based method enhances the function of the nonlinear scaling units by considering some important factors such as the motion simulator's physical constraints and motion sensation error between the drivers in a real vehicle and a motion simulator platform. The proposed method is verified via simulation results which show the superiority of the optimised nonlinear scaling compared with the current trial and error based scaling method for MPC-based MCA as it is able to reduce the sensation error between the motion simulator and real vehicle drivers, enhance motion fidelity, and use the platform workspace more wisely to reduce sensation error while respecting the platform's physical boundaries.

In this paper, we introduce the notion of a monotone fuzzy partition, which is useful for constructing a monotone zero-order Takagi-Sugeno-Kang Fuzzy Inference System (ZOTSK-FIS). It is known that a monotone ZOTSK-FIS model can always be produced when a consistent, complete, and monotone fuzzy rule base is used. However, such an ideal situation is not always available in practice, because a fuzzy rule base is susceptible to uncertainties, e.g., inconsistency, incompleteness, and nonmonotonicity. As a result, we devise an interval method to model these uncertainties by considering the minimum interval of acceptability of a fuzzy rule, resulting in a set of monotone interval-valued fuzzy rules. This further leads to the formulation of a Monotone Interval Fuzzy Inference System (MIFIS) with a minimized uncertainty measure. The proposed MIFIS model is analyzed mathematically and evaluated empirically for the Failure Mode and Effect Analysis (FMEA) application. The results indicate that MIFIS outperforms ZOTSK-FIS, and allows effective decision making using uncertain fuzzy rules solicited from human experts in tackling real-world FMEA problems.

Over the last few decades, pattern classification has become one of the most important fields of artificial intelligence because it constitutes an essential component in many different real-world applications. Artificial neural networks and fuzzy logic are two most widely used models in pattern classification. To build an efficient and powerful model, researchers have introduced hybrid models that combine both fuzzy logic and artificial neural networks. Among the hybrid models, the fuzzy min-max (FMM) neural network has been proven to be a premier model for undertaking pattern classification problems. While FMM is useful in terms of its capability of online learning, it suffers from several limitations in the learning procedure. Therefore, over the past years, researchers have proposed numerous improvements to overcome the limitations of the original FMM model. This paper carries out a comprehensive survey of the developments conducted on the FMM model for pattern classification. In order to assist recent researchers in selecting the most suitable FMM variant and to provide proper guidance for future developments, this study divides the variants of FMM into two main board categories, namely FMM variants with and without contraction. This division facilitates understanding of the developments conducted by researchers on the original FMM neural network, as well as provides the scope to identify the limitations that still exist in the FMM models. This paper also summarizes the use of FMM and its variants in solving different benchmark and real-world problems. Finally, the possible future trends are highlighted.

Autonomous vehicles (AVs) are considered one of the most promising solutions for enhancing road safety, saving individuals' time, and reducing energy consumption. Autonomous vehicles are still in their early stage to be publicly accepted and gain a high level of trust. They need to be comprehensively and continuously evaluated and improved through road tests which are risky, costly, and time-consuming. Motion simulators are capable of contributing to these tests by providing an immersive virtual environment and high fidelity ride experiences for subjective and objective evaluations of AVs' performance. This paper provides a simulation study on the capability of a serial motion platform, known as the Universal Motion Simulator (UMS), for emulation of an AVs' path tracking capabilities. For this purpose, a versatile path tracking model of an AV is initially introduced. The computational Multi-Body System (MBS) approach is then used to implement inverse kinematics and dynamics analyses of the UMS. The UMS is also equipped with an optimal Motion Cueing Algorithm (MCA) to emulate the motion sensation of the AV performing different manoeuvres. The results show that the UMS is an efficient tool for regenerating a realistic and high-fidelity AV ride experience when (1) curvature of the trajectory is not large and the AV does not experience large turning angles when it negotiates the path, and (2) the human motion sensation (vestibular system mathematical model) is taken into consideration in developing the MCA of the motion simulator.

Abstract The bee colony optimization (BCO) algorithm with a linear dance function (denoted as the BCO‐Linear algorithm) is inspired by the bees' foraging behaviors, in which waggle dances are modeled as a communication medium among bees. Through these informative waggle dances, more bees are recruited toward exploring more profitable search regions. In the BCO‐Linear algorithm, a fitter bee is allowed to dance longer, and the dance duration is determined by a linear function with a scaling parameter that requires manual tuning. This article presents a dynamic fuzzy‐based dance mechanism, ie, the BCO‐Fuzzy algorithm, to solve the manual tuning problem. A fuzzy‐based approach is applied to regulate the duration of waggle dances instead of regulating the dance duration using a linear function. The proposed BCO‐Fuzzy algorithm comprises parameters that are dynamically controlled based on the feedback of the search process, therefore overcoming the limitation of manual parameter tuning of the BCO‐Linear algorithm. The BCO‐Fuzzy algorithm is evaluated comprehensively using a set of benchmark traveling salesman problems. The experimental results show that the performance of the BCO‐Fuzzy algorithm is comparable with that of the BCO‐Linear algorithm. Specifically, the dynamic fuzzy‐based dance mechanism improves the BCO algorithm in terms of rewarding dance instances near the inflection point. Performance comparison with other nature‐inspired algorithms proves the effectiveness of the proposed BCO‐Fuzzy algorithm.

Abstract In this paper, an online soft computing model based on an integration between the fuzzy ARTMAP (FAM) neural network and the classification and regression tree (CART) for undertaking data classification problems is presented. Online FAM network is useful for conducting incremental learning with data samples, whereas the CART model prevails in depicting the knowledge learned explicitly in a tree structure. Capitalizing on their respective advantages, the hybrid FAM‐CART model is capable of learning incrementally while explaining its predictions with knowledge elicited from data samples. To evaluate the usefulness of FAM‐CART, 2 sets of benchmark experiments with a total of 12 problems are used in both offline and online learning modes. The results are examined and compared with those published in the literature. The experimental outcome positively indicates that the online FAM‐CART model is useful for tackling data classification tasks. In addition, a decision tree is produced to allow users in understanding the predictions, which is an important property of the hybrid FAM‐CART model in supporting decision‐making tasks.

A system identification-based framework is used to develop monotone fuzzy If-Then rules for formulating monotone zero-order Takagi-Sugeno-Kang (TSK) fuzzy inference systems (FISs) in this paper. Convex and normal trapezoidal and triangular fuzzy sets, together with a strong fuzzy partition strategy (either fixed or adaptive), is adopted. By coupling the strong fuzzy partition with a set of complete and monotone fuzzy If-Then rules, a monotone TSK FIS model can be guaranteed. We show that when a clean multiattribute monotone dataset is used, a system identification-based framework does not guarantee the production of monotone fuzzy If-Then rules, which leads to nonmonotone TSK FIS models. This is a new learning phenomenon that needs to be scrutinized when we design data-based monotone TSK FIS models. Two solutions are proposed: 1) a new monotone fuzzy rule relabeling-based method and 2) a constrained derivative-based optimization method. A new modeling framework with an adaptive fuzzy partition is evaluated. The results indicate that TSK FIS models with better accuracy (a lower sum square error) and a good degree of monotonicity (measured with a monotonicity test) are achieved. In short, the main contributions of this study are validation of the new learning phenomenon and introduction of useful methods for developing data-based monotone TSK FIS models.

This paper presents a novel approach to automatically detect and count cars in different aerial images, which can be satellite or unmanned aerial vehicle (UAV) images. Variations in satellite and/or UAV data make it particularly challenging to have a robust method that works properly on a variety of images. A solution based on the bag-of-words (BoW) model is explored in this paper due to its invariance characteristic and highly stable performance in object/scene categorization. Different from categorization tasks, vehicle detection needs to localize the positions of cars in images. To make BoW suitable for this purpose, we extensively improve the methodology in three aspects, namely, by introducing a recently proposed feature representation, i.e., the local steering kernel descriptor, adding spatial structure constraints, and developing an orientation aware scanning mechanism to produce detection with “one-window-one-car” results. Experiments are conducted on various aerial images with large variations, which consist of data from two public databases, e.g., the Overhead Imagery Research Data Set and Vehicle Detection in Aerial Imagery, as well as other satellite and UAV images. The results demonstrate the effectiveness and robustness of the proposed method. Compared with existing techniques, the proposed method is applicable to a wider range of aerial images.

In this paper, synchronization of an inertial neural network with time-varying delays is investigated. Based on the variable transformation method, we transform the second-order differential equations into the first-order differential equations. Then, using suitable Lyapunov-Krasovskii functionals and Jensen's inequality, the synchronization criteria are established in terms of linear matrix inequalities. Moreover, a feedback controller is designed to attain synchronization between the master and slave models, and to ensure that the error model is globally asymptotically stable. Numerical examples and simulations are presented to indicate the effectiveness of the proposed method. Besides that, an image encryption algorithm is proposed based on the piecewise linear chaotic map and the chaotic inertial neural network. The chaotic signals obtained from the inertial neural network are utilized for the encryption process. Statistical analyses are provided to evaluate the effectiveness of the proposed encryption algorithm. The results ascertain that the proposed encryption algorithm is efficient and reliable for secure communication applications.

In this paper, ad hoc and system identification methods are used to generate fuzzy If-Then rules for a zero-order Takagi-Sugeno-Kang (TSK) Fuzzy Inference System (FIS) using a set of multi-attribute monotone data. Convex and normal trapezoidal fuzzy sets, with a strong fuzzy partition strategy, is employed. Our analysis shows that even with multi-attribute monotone data, non-monotone fuzzy If-Then rules can be produced using an ad hoc method. The same observation can be made, empirically, using a system identification method, e.g., a derivative-based optimization method and the genetic algorithm. This finding is important for modeling a monotone FIS model, as the result shows that even with a “clean” data set pertaining to a monotone system, the generated fuzzy If-Then rules may need to be pre-processed, before being used for FIS modeling. As such, monotone fuzzy rule relabeling is useful. Besides that, a constrained non-linear programming method for FIS modelling is suggested, as a variant of the system identification method.

Randomized neural network (RNN) is a highly feasible solution in the era of big data because it offers a simple and fast working principle in processing dynamic and evolving data streams. This paper proposes a novel RNN, namely recurrent type-2 random vector functional link network (RT2McRVFLN), which provides a highly scalable solution for data streams in a strictly online and integrated framework. It is built upon the psychologically inspired concept of metacognitive learning, which covers three basic components of human learning: what-to-learn, how-to-learn, and when-to-learn. The what-to-learn selects important samples on the fly with the use of online active learning scenario, which renders our algorithm an online semi-supervised algorithm. The how-to-learn process combines an open structure of evolving concept and a randomized learning algorithm of random vector functional link network (RVFLN). The efficacy of the RT2McRVFLN has been numerically validated through two real-world case studies and comparisons with its counterparts, which arrive at a conclusive finding that our algorithm delivers a tradeoff between accuracy and simplicity.

The Artificial Bee Colony (ABC) algorithm is a swarm intelligence approach which has initially been proposed to solve optimization of mathematical test functions with a unique neighbourhood search mechanism. However, this neighbourhood search mechanism could not be directly applied to combinatorial discrete optimization problems. The employed and onlooker bees need to be equipped with problem-specific perturbative heuristics in order to tackle combinatorial discrete optimization problems. However, there is a large variety of available problem-specific heuristics. In this paper, a hyper-heuristic method, namely a Modified Choice Function (MCF), is applied such that it can regulate the selection of the neighbourhood search heuristics adopted by the employed and onlooker bees automatically. The proposed MCF-based ABC model is implemented using the Hyper-heuristic Flexible Framework (HyFlex). To demonstrate the effectiveness of the proposed model, ten Traveling Salesman Problem (TSP) instances available in HyFlex have been evaluated. The empirical results show that the proposed model is able to statistically outperform four out of five ABC variants throughout the optimization process.

This paper proposes a facial expression recognition system using evolutionary particle swarm optimization (PSO)-based feature optimization. The system first employs modified local binary patterns, which conduct horizontal and vertical neighborhood pixel comparison, to generate a discriminative initial facial representation. Then, a PSO variant embedded with the concept of a micro genetic algorithm (mGA), called mGA-embedded PSO, is proposed to perform feature optimization. It incorporates a nonreplaceable memory, a small-population secondary swarm, a new velocity updating strategy, a subdimension-based in-depth local facial feature search, and a cooperation of local exploitation and global exploration search mechanism to mitigate the premature convergence problem of conventional PSO. Multiple classifiers are used for recognizing seven facial expressions. Based on a comprehensive study using within- and cross-domain images from the extended Cohn Kanade and MMI benchmark databases, respectively, the empirical results indicate that our proposed system outperforms other state-of-the-art PSO variants, conventional PSO, classical GA, and other related facial expression recognition models reported in the literature by a significant margin.

This brief is mainly concerned with a series of dynamical analyses of the Hindmarsh-Rose (HR) neuron with state-dependent time delays. The dynamical analyses focus on stability, Hopf bifurcation, as well as chaos and chaos control. Through the stability and bifurcation analysis, we determine that increasing the external current causes the excitable HR neuron to exhibit periodic or chaotic bursting/spiking behaviors and emit subcritical Hopf bifurcation. Furthermore, by choosing a fixed external current and varying the time delay, the stability of the HR neuron is affected. We analyze the chaotic behaviors of the HR neuron under a fixed external current through time series, bifurcation diagram, Lyapunov exponents, and Lyapunov dimension. We also analyze the synchronization of the chaotic time-delayed HR neuron through nonlinear control. Based on an appropriate Lyapunov-Krasovskii functional with triple integral terms, a nonlinear feedback control scheme is designed to achieve synchronization between the uncontrolled and controlled models. The proposed synchronization criteria are derived in terms of linear matrix inequalities to achieve the global asymptotical stability of the considered error model under the designed control scheme. Finally, numerical simulations pertaining to stability, Hopf bifurcation, periodic, chaotic, and synchronized models are provided to demonstrate the effectiveness of the derived theoretical results.

The aim of this paper is to design and develop an optimal motion cueing algorithm (MCA) based on the genetic algorithm (GA) that can generate high-fidelity motions within the motion simulator's physical limitations. Both, angular velocity and linear acceleration are adopted as the inputs to the MCA for producing the higher order optimal washout filter. The linear quadratic regulator (LQR) method is used to constrain the human perception error between the real and simulated driving tasks. To develop the optimal MCA, the latest mathematical models of the vestibular system and simulator motion are taken into account. A reference frame with the center of rotation at the driver's head to eliminate false motion cues caused by rotation of the simulator to the translational motion of the driver's head as well as to reduce the workspace displacement is employed. To improve the developed LQR-based optimal MCA, a new strategy based on optimal control theory and the GA is devised. The objective is to reproduce a signal that can follow closely the reference signal and avoid false motion cues by adjusting the parameters from the obtained LQR-based optimal washout filter. This is achieved by taking a series of factors into account, which include the vestibular sensation error between the real and simulated cases, the main dynamic limitations, the human threshold limiter in tilt coordination, the cross correlation coefficient, and the human sensation error fluctuation. It is worth pointing out that other related investigations in the literature normally do not consider the effects of these factors. The proposed optimized MCA based on the GA is implemented using the MATLAB/Simulink software. The results show the effectiveness of the proposed GA-based method in enhancing human sensation, maximizing the reference shape tracking, and reducing the workspace usage.

The use of Fuzzy Inference System (FIS) in decision making problems has received little attention so far. This may be due to the difficulty in gathering a complete set of fuzzy rules, which is free from noise, and the complexity in constructing an FIS model that is able to satisfy a number of important properties, including the monotonicity property. Previously, we have proposed a single-input Monotone-Interval FIS (MI-FIS) model, which can handle incomplete and non-monotone fuzzy rules. Besides that, we have proposed the idea of a monotone test (MT) for a set of fuzzy rules, which give an indication pertaining to the degree of monotonicity of a fuzzy rules set. In this paper, a multi-input MI-FIS model is firstly presented. The focus of this paper is on the use of MI-FIS and MT for undertaking multi expert decision-making (MEDM) problems. A three-phase MEDM framework consists of modelling, aggregation, and exploitation phases is proposed. In the modelling phase, an MT index for each fuzzy rule base from each expert, which is potentially non-monotone and incomplete, is obtained. The provided fuzzy rule bases are also modelled as MI-FISs. In the aggregation phase, an overall collective rating score of an alternative from a number of experts is obtained through the fuzzy weighted averaging operator. We suggest including MT as part of the aggregation phase. In exploitation phase, a rank ordering procedure among the alternatives is established using a possibility method. The developed framework is evaluated with simulated information. The results show that including the MT index in the aggregation phase is able to increase the robustness of the proposed FIS-MEDM model in the presence of noisy fuzzy rule sets.

The washout filter for a driving simulator is able to regenerate high fidelity vehicle translational and rotational motions within the simulator's physical limitations and return the simulator platform back to its initial position. The classical washout filter provides a popular solution that has been broadly utilized in different commercial simulators due to its simplicity, short processing time, and reasonable performance. One limitation of the classical washout filter is its sub-optimal parameter tuning process, which is based on the trial-and-error method. This leads to an inefficient workspace usage and, consequently, generation of false motion cues that lead to simulator sickness. Ignorance of a human sensation model in its design is another drawback of classical washout filters. The purpose of this study is to use Particle Swarm Optimization (PSO) to design and tune the washout filter parameters, in order to increase motion fidelity, decrease the human sensation error, and improve efficiency of the workspace usage. The proposed PSO-based washout filter is designed and implemented using the MATLAB/Simulink software package. The results indicate the effectiveness of the PSO-based washout filter in reducing the human sensation error, increasing the capability of reference shape tracking, and improving efficiency of the workspace usage.

Recent efforts in the evolutionary multi-objective optimization (EMO) community focus on addressing shortcomings of current solution techniques adopted for solving many-objective optimization problems (MaOPs). One such challenge faced by classical multi-objective evolutionary algorithms is diversity preservation in optimization problems with more than three objectives, namely MaOPs. In this vein, NSGA-III has replaced the crowding distance measure in NSGA-II with reference points in the objective space to ensure diversity of the converged solutions along the pre-determined solutions in the environmental selection phase. NSGA-III uses the Pareto-dominance principle to obtain the non-dominated solutions in the environmental selection phase. However, the Pareto-dominance principle loses its selection pressure in high-dimensional optimization problems, because most of the obtained solutions become non-dominated. Inspired by θ-DEA, we address the selection pressure issue in NSGA-III, by exploiting the decomposition principle of MOEA/D using reference points for multiple single-objective optimization problems. Moreover, similar to MOEA/D, the parent selection process is restricted to the neighboring solutions, as opposed to random selection of parent solutions from the entire population in NSGA-III. The effectiveness of the proposed method is demonstrated on different well-known benchmark optimization problems for 3- to 10-objectives. The results compare favorably with those from MOEA/D, NSGA-III, and θ-DEA.

In this research, we propose a facial expression recognition system with a variant of evolutionary firefly algorithm for feature optimization. First of all, a modified Local Binary Pattern descriptor is proposed to produce an initial discriminative face representation. A variant of the firefly algorithm is proposed to perform feature optimization. The proposed evolutionary firefly algorithm exploits the spiral search behaviour of moths and attractiveness search actions of fireflies to mitigate premature convergence of the Levy-flight firefly algorithm (LFA) and the moth-flame optimization (MFO) algorithm. Specifically, it employs the logarithmic spiral search capability of the moths to increase local exploitation of the fireflies, whereas in comparison with the flames in MFO, the fireflies not only represent the best solutions identified by the moths but also act as the search agents guided by the attractiveness function to increase global exploration. Simulated Annealing embedded with Levy flights is also used to increase exploitation of the most promising solution. Diverse single and ensemble classifiers are implemented for the recognition of seven expressions. Evaluated with frontal-view images extracted from CK+, JAFFE, and MMI, and 45-degree multi-view and 90-degree side-view images from BU-3DFE and MMI, respectively, our system achieves a superior performance, and outperforms other state-of-the-art feature optimization methods and related facial expression recognition models by a significant margin.

To maintain the monotonicity property of a fuzzy inference system, a monotonically ordered and complete set of fuzzy rules is necessary. However, monotonically ordered fuzzy rules are not always available, e.g., errors in human judgments lead to nonmonotone fuzzy rules. The focus of this paper is on a new monotone fuzzy rule relabeling (MFRR) method that is able to relabel a set of nonmonotone fuzzy rules to meet the monotonicity property with reduced computation. Unlike the brute-force approach, which is susceptible to the combinatorial explosion problem, the proposed MFRR method explores within a reduced search space to find the solutions, therefore decreasing the computational requirements. The usefulness of the proposed method in undertaking failure mode and effect analysis problems is demonstrated using publicly available information. The results indicate that the MFRR method can produce optimal solutions with reduced computational time.

A medical case study related to implantable rotary blood pumps is examined. Five classifiers and two ensemble classifiers are applied to process the signals collected from the pumps for the identification of the aortic valve nonopening pump state. In addition to the noise-free datasets, up to 40% class noise has been added to the signals to evaluate the classification performance when mislabeling is present in the classifier training set. In order to ensure a reliable diagnostic model for the identification of the pump states, classifications performed with and without class noise are evaluated. The multilayer perceptron emerged as the best performing classifier for pump state detection due to its high accuracy as well as robustness against class noise.

A hybrid intelligent model comprising a modified fuzzy min-max (FMM) clustering neural network and a modified clustering tree (CT) is developed. A review of clustering models with rule extraction capabilities is presented. The hybrid FMM-CT model is explained. We first use several benchmark problems to illustrate the cluster evolution patterns from the proposed modifications in FMM. Then, we employ a case study with real data related to power quality monitoring to assess the usefulness of FMM-CT. The results are compared with those from other clustering models. More importantly, we extract explanatory rules from FMM-CT to justify its predictions. The empirical findings indicate the usefulness of the proposed model in tackling data clustering and power quality monitoring problems under different environments.

Named Entity Recognition (NER) is a crucial step in text mining. This paper proposes a new graph-based technique for representing unstructured medical text. The new representation is used to extract discriminative features that are able to enhance the NER performance. To evaluate the usefulness of the proposed graph-based technique, the i2b2 medication challenge data set is used. Specifically, the 'treatment' named entities are extracted for evaluation using six different classifiers. The F-measure results of five classifiers are enhanced, with an average improvement of up to 26% in performance.

An enhanced fuzzy min-max (EFMM) network is proposed for pattern classification in this paper. The aim is to overcome a number of limitations of the original fuzzy min-max (FMM) network and improve its classification performance. The key contributions are three heuristic rules to enhance the learning algorithm of FMM. First, a new hyperbox expansion rule to eliminate the overlapping problem during the hyperbox expansion process is suggested. Second, the existing hyperbox overlap test rule is extended to discover other possible overlapping cases. Third, a new hyperbox contraction rule to resolve possible overlapping cases is provided. Efficacy of EFMM is evaluated using benchmark data sets and a real medical diagnosis task. The results are better than those from various FMM-based models, support vector machine-based, Bayesian-based, decision tree-based, fuzzy-based, and neural-based classifiers. The empirical findings show that the newly introduced rules are able to realize EFMM as a useful model for undertaking pattern classification problems.

This paper presents a new Fuzzy Inference System (FIS)-based Risk Priority Number (RPN) model for the prioritization of failures in Failure Mode and Effect Analysis (FMEA). In FMEA, the monotonicity property of the RPN scores is important. To maintain the monotonicity property of an FIS-based RPN model, a complete and monotonically-ordered fuzzy rule base is necessary. However, it is impractical to gather all (potentially a large number of) fuzzy rules from FMEA users. In this paper, we introduce a new two-stage approach to reduce the number of fuzzy rules that needs to be gathered, and to satisfy the monotonicity property. In stage-1, a Genetic Algorithm (GA) is used to search for a small set of fuzzy rules to be gathered from FMEA users. In stage-2, the remaining fuzzy rules are deduced approximately by a monotonicity-preserving similarity reasoning scheme. The monotonicity property is exploited as additional qualitative information for constructing the FIS-based RPN model. To assess the effectiveness of the proposed approach, a real case study with information collected from a semiconductor manufacturing plant is conducted. The outcomes indicate that the proposed approach is effective in developing an FIS-based RPN model with only a small set of fuzzy rules, which is able to satisfy the monotonicity property for prioritization of failures in FMEA.

In this paper, a new fuzzy ranking method for both type-1 and interval type-2 fuzzy sets (FSs) using fuzzy preference relations is proposed. The use of fuzzy preference relations to rank FSs with vertices has been introduced, and successfully implemented to undertake fuzzy multiple criteria hierarchical group decision-making problems. The proposed fuzzy ranking method is an extension of the results published in [1], and it is able to rank FSs with and without vertices. Besides that, it is important for a fuzzy ranking method to satisfy six reasonable fuzzy ordering properties as discussed in [6]-[8]. As a result, the capability of the proposed fuzzy ranking method in fulfilling these properties is analyzed and discussed. Issues related to time complexity of the proposed method are also examined.

In our previous investigations, two Similarity Reasoning (SR)-based frameworks for tackling real-world problems have been proposed. In both frameworks, SR is used to deduce unknown fuzzy rules based on similarity of the given and unknown fuzzy rules for building a Fuzzy Inference System (FIS). In this paper, we further extend our previous findings by developing (1) a multi-objective evolutionary model for fuzzy rule selection; and (2) an evidential function to facilitate the use of both frameworks. The Non-Dominated Sorting Genetic Algorithms-П (NSGA-П) is adopted for fuzzy rule selection, in accordance with the Pareto optimal criterion. Besides that, two new evidential functions are developed, whereby given fuzzy rules are considered as evidence. Simulated and benchmark examples are included to demonstrate the applicability of these suggestions. Positive results were obtained.

The principle of ratios has been applied to many real world problems, e.g. the part-to-part and part-to-whole ratio formulations. As it is difficult for humans to provide an exact ratio in many real situations, we introduce a fuzzy ratio in this paper. We use some notions from fuzzy arithmetic to analyze fuzzy ratios captured from humans. An application of the formulated fuzzy ratio to a Single Input Rule Modules connected Fuzzy Inference System (SIRMs-FIS) is demonstrated. Instead of using a precise weight, fuzzy sets are employed to represent the relative importance of each rule module. The resulting fuzzy weights are explained as a fuzzy ratio on a weight domain. In addition, a new SIRMs-FIS model with fuzzy weights and part-to-whole fuzzy ratio is devised. A simulated example is presented to clarify the proposed SIRM-FIS model.

The focus of this paper is on handling non-monotone information in the modelling process of a single-input target monotone system. On one hand, the monotonicity property is a piece of useful prior (or additional) information which can be exploited for modelling of a monotone target system. On the other hand, it is difficult to model a monotone system if the available information is not monotonically-ordered. In this paper, an interval-based method for analysing non-monotonically ordered information is proposed. The applicability of the proposed method to handling a non-monotone function, a non-monotone data set, and an incomplete and/or non-monotone fuzzy rule base is presented. The upper and lower bounds of the interval are firstly defined. The region governed by the interval is explained as a coverage measure. The coverage size represents uncertainty pertaining to the available information. The proposed approach constitutes a new method to transform non-monotonic information to interval-valued monotone system. The proposed interval-based method to handle an incomplete and/or non-monotone fuzzy rule base constitutes a new fuzzy reasoning approach.

A search in the literature reveals that mathematical conditions (usually sufficient conditions) for the Fuzzy Inference System (FIS) models to satisfy the monotonicity property have been developed. A monotonically-ordered fuzzy rule base is important to maintain the monotonicity property of an FIS. However, it may difficult to obtain a monotonically-ordered fuzzy rule base in practice. We have previously introduced the idea of fuzzy rule relabeling to tackle this problem. In this paper, we further propose a monotonicity index for the FIS system, which serves as a metric to indicate the degree of a fuzzy rule base fulfilling the monotonicity property. The index is useful to provide an indication whether a fuzzy rule base should (or should not) be used in practice, even with fuzzy rule relabeling. To illustrate the idea, the zero-order Sugeno FIS model is exemplified. We add noise as errors into the fuzzy rule base to formulate a set of non-monotone fuzzy rules. As such, the metric also acts as a measure of noise in the fuzzy rule base. The results show that the proposed metric is useful to indicate the degree of a fuzzy rule base fulfilling the monotonicity property.

A recent study in Science indicated that the confidence of a decision maker played an essential role in group decision making problems. In order to make use of the information of each individual's confidence of the current decision problem, a new hybrid weighted aggregation method to solve a group decision making peoblem is proposed in this paper. Specifically, the hybrid weight of each expert is generated by a convex combination of his/her subjective experience-based weight and objective problem-domain-based weight. The experience-based weight is derived from the expert's historical experiences and the problem-domain-based weight is characterized by the confidence degree and consensus degree of each expert's opinions in the current decision making process. Based on the hybrid weighted aggregation method, all the experts' opinions which are expressed in the form of fuzzy preference relations are consequently aggregated to obtain a collective group opinion. Some valuable properities of the proposed method are discussed. A nurse manager hiring problem in a hospital is employed to illustrate that the proposed method provides a rational and valid solution for the group decision making problem when the experts are not willing to change their initial preferences, or the cost of change is high due to time limitation.

An accurate Named Entity Recognition (NER) is important for knowledge discovery in text mining. This paper proposes an ensemble machine learning approach to recognise Named Entities (NEs) from unstructured and informal medical text. Specifically, Conditional Random Field (CRF) and Maximum Entropy (ME) classifiers are applied individually to the test data set from the i2b2 2010 medication challenge. Each classifier is trained using a different set of features. The first set focuses on the contextual features of the data, while the second concentrates on the linguistic features of each word. The results of the two classifiers are then combined. The proposed approach achieves an f-score of 81.8%, showing a considerable improvement over the results from CRF and ME classifiers individually which achieve f-scores of 76% and 66.3% for the same data set, respectively.

In this paper, the application of a hybrid model combining the fuzzy min-max (FMM) neural network and the classification and regression tree (CART) to human activity recognition is presented. The hybrid FMM-CART model capitalizes the merits of both FMM and CART in data classification and rule extraction. To evaluate the effectiveness of FMM-CART, two data sets related to human activity recognition problems are conducted. The results obtained are higher than those reported in the literature. More importantly, practical rules in the form of a decision tree are extracted to provide explanation and justification for the predictions from FMM-CART. This outcome positively indicates the potential of FMM-CART in undertaking human activity recognition tasks.

In this brief, a hybrid model combining the fuzzy min-max (FMM) neural network and the classification and regression tree (CART) for online motor detection and diagnosis tasks is described. The hybrid model, known as FMM-CART, exploits the advantages of both FMM and CART for undertaking data classification and rule extraction problems. To evaluate the applicability of the proposed FMM-CART model, an evaluation with a benchmark data set pertaining to electrical motor bearing faults is first conducted. The results obtained are equivalent to those reported in the literature. Then, a laboratory experiment for detecting and diagnosing eccentricity faults in an induction motor is performed. In addition to producing accurate results, useful rules in the form of a decision tree are extracted to provide explanation and justification for the predictions from FMM-CART. The experimental outcome positively shows the potential of FMM-CART in undertaking online motor fault detection and diagnosis tasks.

In this paper, a new Fuzzy Set (FS) ranking method (for type-1 and interval type-2 FSs), which is based on the Dempster-Shafer Theory (DST) of evidence with fuzzy targets, is investigated. Fuzzy targets are adopted to reflect human viewpoints on fuzzy ranking. Two important measures in DST, i.e., the belief and plausibility measures, are used to rank FSs. The proposed approach is evaluated with several benchmark examples. The use of the belief and plausibility measures in fuzzy ranking are discussed and compared. We further analyze the capability of the proposed approach in fulfilling six reasonable fuzzy ordering properties as discussed in [9]-[11].

An improved evolving model, i.e., Evolving Tree (ETree) with Fuzzy c-Means (FCM), is proposed for undertaking text document visualization problems in this study. ETree forms a hierarchical tree structure in which nodes (i.e., trunks) are allowed to grow and split into child nodes (i.e., leaves), and each node represents a cluster of documents. However, ETree adopts a relatively simple approach to split its nodes. Thus, FCM is adopted as an alternative to perform node splitting in ETree. An experimental study using articles from a flagship conference of Universiti Malaysia Sarawak (UNIMAS), i.e., Engineering Conference (ENCON), is conducted. The experimental results are analyzed and discussed, and the outcome shows that the proposed ETree-FCM model is effective for undertaking text document clustering and visualization problems.

In this paper, a new online updating framework for constructing monotonicity-preserving Fuzzy Inference Systems (FISs) is proposed. The framework encompasses an optimization-based Similarity Reasoning (SR) scheme and a new monotone fuzzy rule relabeling technique. A complete and monotonically-ordered fuzzy rule base is necessary to maintain the monotonicity property of an FIS model. The proposed framework attempts to allow a monotonicity-preserving FIS model to be constructed when the fuzzy rules are incomplete and not monotonically-ordered. An online feature is introduced to allow the FIS model to be updated from time to time. We further investigate three useful measures, i.e., the belief, plausibility, and evidential mass measures, which are inspired from the Dempster-Shafer theory of evidence, to analyze the proposed framework and to give an insight for the inferred outcomes from the FIS model.

A complete and monotonically-ordered fuzzy rule base is necessary to maintain the monotonicity property of a Fuzzy Inference System (FIS). In this paper, a new monotone fuzzy rule relabeling technique to relabel a non-monotone fuzzy rule base provided by domain experts is proposed. Even though the Genetic Algorithm (GA)-based monotone fuzzy rule relabeling technique has been investigated in our previous work [7], the optimality of the approach could not be guaranteed. The new fuzzy rule relabeling technique adopts a simple brute force search, and it can produce an optimal result. We also formulate a new two-stage framework that encompasses a GA-based rule selection scheme, the optimization based-Similarity Reasoning (SR) scheme, and the proposed monotone fuzzy rule relabeling technique for preserving the monotonicity property of the FIS model. Applicability of the two-stage framework to a real world problem, i.e., failure mode and effect analysis, is further demonstrated. The results clearly demonstrate the usefulness of the proposed framework.

Even though the importance of the local monotonicity property for function approximation problems is well established, there are relative few investigations addressing issues related to the fulfillment of the local monotonicity property in Fuzzy Inference System (FIS) modeling. We have previously conducted a preliminary study on the local monotonicity property of FIS models, with the assumption that the extrema point(s) (i.e., the maximum and/or minimum point(s)) is either known precisely or totally unknown. However, in some practical situations, the extrema point(s) can be known imprecisely (as an interval or a fuzzy set). In this paper, the imprecise information is exploited to construct an FIS model that fulfills the local monotonicity property. A procedure to estimate the extrema point(s) of a function is devised. Applicability of the findings to a data-driven modeling problem is further demonstrated.

The understanding of cell manipulation, for example in microinjection, requires an accurate model of the cells. Motivated by this important requirement, a 3D particle-based mechanical model is derived for simulating the deformation of the fish egg membrane and the corresponding cellular forces during micro robotic cell injection. The model is formulated based on the kinematic and dynamic of spring-damper configuration with multi-particle joints considering the visco-elastic fluidic properties. It simulates the indentation force feedback as well as cell visual deformation during microinjection. A preliminary simulation study is conducted with different parameter configurations. The results indicate that the proposed particle-based model is able to provide similar deformation profiles as observed from a real microinjection experiment of the zebra fish embryo published in the literature. As a generic modelling approach is adopted, the proposed model also has the potential in applications with different types of manipulation such as micropipette cell aspiration.

In this paper, a novel approach to building a Fuzzy Inference System (FIS) that preserves the monotonicity property is proposed. A new fuzzy re-labeling technique to re-label the consequents of fuzzy rules in the database (before the Similarity Reasoning process) and a monotonicity index for use in FIS modeling are introduced. The proposed approach is able to overcome several restrictions in our previous work that uses mathematical conditions in building monotonicity-preserving FIS models. Here, we show that the proposed approach is applicable to different FIS models, which include the zero-order Sugeno FIS and Mamdani models. Besides, the proposed approach can be extended to undertake problems related to the local monotonicity property of FIS models. A number of examples to demonstrate the usefulness of the proposed approach are presented. The results indicate the usefulness of the proposed approach in constructing monotonicity-preserving FIS models.

In this paper, the problem of maintaining the (global) monotonicity and local monotonicity properties between the input(s) and the output of an FIS model is addressed. This is known as the monotone fuzzy modeling problem. In our previous work, this problem has been tackled by developing some mathematical conditions for an FIS model to observe the monotonicity property. These mathematical conditions are used as a set of governing equations for undertaking FIS modeling problems, and have been extended to some advanced FIS modeling techniques. Here, we examine an alternative to the monotone fuzzy modeling problem by introducing a monotonicity index. The monotonicity index is employed as an approximate indicator to measure the fulfillment of an FIS model to the monotonicity property. It allows the FIS model to be constructed using an optimization method, or be tuned to achieve a better performance, without knowing the exact mathematical conditions of the FIS model to satisfy the monotonicity property. Besides, the monotonicity index can be extended to FIS modeling that involves the local monotonicity problem. We also analyze the relationship between the FIS model and its monotonicity property fulfillment, as well as derived mathematical conditions, using the Monte Carlo method.

In this paper, a novel approach to detect and classify comprehensive fault conditions of induction motors using a hybrid fuzzy min-max (FMM) neural network and classification and regression tree (CART) is proposed. The hybrid model, known as FMM-CART, exploits the advantages of both FMM and CART for undertaking data classification and rule extraction problems. A series of real experiments is conducted, whereby the motor current signature analysis method is applied to form a database comprising stator current signatures under different motor conditions. The signal harmonics from the power spectral density are extracted as discriminative input features for fault detection and classification with FMM-CART. A comprehensive list of induction motor fault conditions, viz., broken rotor bars, unbalanced voltages, stator winding faults, and eccentricity problems, has been successfully classified using FMM-CART with good accuracy rates. The results are comparable, if not better, than those reported in the literature. Useful explanatory rules in the form of a decision tree are also elicited from FMM-CART to analyze and understand different fault conditions of induction motors.

In this paper, an Evolutionary-based Similarity Reasoning (ESR) scheme for preserving the monotonicity property of the multi-input Fuzzy Inference System (FIS) is proposed. Similarity reasoning (SR) is a useful solution for undertaking the incomplete rule base problem in FIS modeling. However, SR may not be a direct solution to designing monotonic multi-input FIS models, owing to the difficulty in getting a set of monotonically-ordered conclusions. The proposed ESR scheme, which is a synthesis of evolutionary computing, sufficient conditions, and SR, provides a useful solution to modeling and preserving the monotonicity property of multi-input FIS models. A case study on Failure Mode and Effect Analysis (FMEA) is used to demonstrate the effectiveness of the proposed ESR scheme in undertaking real world problems that require the monotonicity property of FIS models.

In this paper, two issues relating to modeling of a monotonicity-preserving Fuzzy Inference System (FIS) are examined. The first is on designing or tuning of Gaussian Membership Functions (MFs) for a monotonic FIS. Designing Gaussian MFs for an FIS is difficult because of its spreading and curvature characteristics. In this study, the sufficient conditions are exploited, and the procedure of designing Gaussian MFs is formulated as a constrained optimization problem. The second issue is on the testing procedure for a monotonic FIS. As such, a testing procedure for a monotonic FIS model is proposed. Applicability of the proposed approach is demonstrated with a real world industrial application, i.e., Failure Mode and Effect Analysis. The results obtained are analysis and discussed. The outcomes show that the proposed approach is useful in designing a monotonicity-preserving FIS model.

An important and difficult issue in designing a Fuzzy Inference System (FIS) is the specification of fuzzy sets and fuzzy rules. In this paper, two useful qualitative properties of the FIS model, i.e., the monotonicity and sub-additivity properties, are studied. The monotonic sufficient conditions of the FIS model with Gaussian membership functions are further analyzed. The aim is to incorporate the sufficient conditions into the FIS modeling process, which serves as a simple (which can be easily understood by domain users), easy-to-use (which can be easily applied to or can be a part of the FIS model), and yet reliable (which has a sound mathematical foundation) method to preserve the monotonicity property of the FIS model. Another aim of this paper is to demonstrate how these additional qualitative information can be exploited and extended to be part of the FIS designing procedure (i.e., for fuzzy sets and fuzzy rules design) via the sufficient conditions (which act as a set of useful governing equations for designing the FIS model). The proposed approach is able to avoid the "trial and error" procedure in obtaining a monotonic FIS model. To assess the applicability of the proposed approach, two practical problems are examined. The first is an FIS-based model for water level control, while the second is an FIS-based Risk Priority Number (RPN) model in Failure Mode and Effect Analysis (FMEA). To further illustrate the importance of the sufficient conditions as the governing equations, an analysis on the consequences of violating the sufficient conditions of the FIS-based RPN model is presented.

Generalized adaptive resonance theory (GART) is a neural network model that is capable of online learning and is effective in tackling pattern classification tasks. In this paper, we propose an improved GART model (IGART), and demonstrate its applicability to power systems. IGART enhances the dynamics of GART in several aspects, which include the use of the Laplacian likelihood function, a new vigilance function, a new match-tracking mechanism, an ordering algorithm for determining the sequence of training data, and a rule extraction capability to elicit if-then rules from the network. To assess the effectiveness of IGART and to compare its performances with those from other methods, three datasets that are related to power systems are employed. The experimental results demonstrate the usefulness of IGART with the rule extraction capability in undertaking classification problems in power systems engineering.

An important and difficult issue in designing a Fuzzy Inference System (FIS) is the specification of fuzzy sets, and fuzzy rules. The aim of this paper is to demonstrate how an additional qualitative information, i.e., monotonicity property, can be exploited and extended to be part of an FIS designing procedure (i.e., fuzzy sets and fuzzy rules design). In this paper, the FIS is employed as an alternative to the use of addition in aggregating the scores from test items/tasks in a Criterion-Referenced Assessment (CRA) model. In order to preserve the monotonicity property, the sufficient conditions of the FIS is proposed. Our proposed FIS based CRA procedure can be viewed as an enhancement for the FIS based CRA procedure, where monotonicity property is preserved. We demonstrate the applicability of the proposed approach with a case study related to a laboratory project assessment task at a university, and the results indicate the usefulness of the proposed approach in the CRA domain.

High-mix-low-volume (HMLV) production is currently a worldwide manufacturing trend. It requires a high degree of customization in the manufacturing process to produce a wide range of products in low quantity in order to meet customers' demand for more variety and choices of products. Such a kind of business environment has increased the conversion time and decreased the production efficiency due to frequent production changeover. In this paper, a layered-encoding cascade optimization (LECO) approach is proposed to develop an HMLV product-mix optimizer that exhibits the benefits of low conversion time, high productivity, and high equipment efficiency. Specifically, the genetic algorithm (GA) and particle swarm optimization (PSO) techniques are employed as optimizers for different decision layers in different LECO models. Each GA and PSO optimizer is studied and compared. A number of hypothetical and real data sets from a manufacturing plant are used to evaluate the performance of the proposed GA and PSO optimizers. The results indicate that, with a proper selection of the GA and PSO optimizers, the LECO approach is able to generate high-quality product-mix plans to meet the production demands in HMLV manufacturing environments.

In this paper, a two-stage pattern classification and rule extraction system is proposed. The first stage consists of a modified fuzzy min-max (FMM) neural-network-based pattern classifier, while the second stage consists of a genetic-algorithm (GA)-based rule extractor. Fuzzy if-then rules are extracted from the modified FMM classifier, and a ¿don't care¿ approach is adopted by the GA rule extractor to minimize the number of features in the extracted rules. Five benchmark problems and a real medical diagnosis task are used to empirically evaluate the effectiveness of the proposed FMM-GA system. The results are analyzed and compared with other published results. In addition, the bootstrap hypothesis analysis is conducted to quantify the results of the medical diagnosis task statistically. The outcomes reveal the efficacy of FMM-GA in extracting a set of compact and yet easily comprehensible rules while maintaining a high classification performance for tackling pattern classification tasks.

Constructing a monotonicity relating function is important, as many engineering problems revolve around a monotonicity relationship between input(s) and output(s). In this paper, we investigate the use of fuzzy rule interpolation techniques for monotonicity relating fuzzy inference system (FIS). A mathematical derivation on the conditions of an FIS to be monotone is provided. From the derivation, two conditions are necessary. The derivation suggests that the mapped consequence fuzzy set of an FIS to be of a monotonicity order. We further evaluate the use of fuzzy rule interpolation techniques in predicting a consequent associated with an observation according to the monotonicity order. There are several findings in this article. We point out the importance of an ordering criterion in rule selection for a multi-input FIS before the interpolation process; and hence, the practice of choosing the nearest rules may not be true in this case. To fulfill the monotonicity order, we argue with an example that conventional fuzzy rule interpolation techniques that predict each consequence separately is not suitable in this case. We further suggest another class of interpolation techniques that predicts the consequence of a set of observations simultaneously, instead of separately. This can be accomplished with the use of a search algorithm, such as the brute force, genetic algorithm or etc.

In this paper, an interactive genetic algorithm (IGA) approach is developed to optimize design variables for a monolithic microwave integrated circuit (MMIC) low noise amplifier. A layered encoding structure is employed to the problem representation in genetic algorithm to allow human intervention in the circuit design variable tuning process. The MMIC amplifier design is synthesized using the Agilent Advance Design System (ADS), and the IGA is proposed to tune the design variables in order to meet multiple constraints and objectives such as noise figure, current and simulated power gain. The developed IGA is compared with other optimization techniques from ADS. The results showed that the IGA performs better in achieving most of the involved objectives.

This paper presents a fuzzy ARTMAP (FAM) based modular architecture for multi-class pattern recognition known as modular adaptive resonance theory map (MARTMAP). The prediction of class membership is made collectively by combining outputs from multiple novelty detectors. Distance-based familiarity discrimination is introduced to improve the robustness of MARTMAP in the presence of noise. The effectiveness of the proposed architecture is analyzed and compared with ARTMAP-FD network, FAM network, and One-Against-One Support Vector Machine (OAO-SVM). Experimental results show that MARTMAP is able to retain effective familiarity discrimination in noisy environment, and yet less sensitive to class imbalance problem as compared to its counterparts.

In this brief, a new neural network model called generalized adaptive resonance theory (GART) is introduced. GART is a hybrid model that comprises a modified Gaussian adaptive resonance theory (MGA) and the generalized regression neural network (GRNN). It is an enhanced version of the GRNN, which preserves the online learning properties of adaptive resonance theory (ART). A series of empirical studies to assess the effectiveness of GART in classification, regression, and time series prediction tasks is conducted. The results demonstrate that GART is able to produce good performances as compared with those of other methods, including the online sequential extreme learning machine (OSELM) and sequential learning radial basis function (RBF) neural network models.

A new lattice disentangling monitoring algorithm for a hybrid self-organizing map-kernel-based maximum entropy learning rule (SOM-kMER) model is proposed. It aims to overcome topological defects owing to a rapid decrease of the neighborhood range over the finite running time in topographic map formation. The empirical results demonstrate that the proposed approach is able to accelerate the formation of a topographic map and, at the same time, to simplify the monitoring procedure.

A hybrid neural network comprising Fuzzy ARTMAP and Fuzzy C-Means Clustering is proposed for pattern classification with incomplete training and test data. Two benchmark problems and a real medical pattern classification task are employed to evaluate the effectiveness of the hybrid network. The results are analyzed and compared with those from other methods.

A hybrid neural network model, based on the fusion of fuzzy adaptive resonance theory (FA ART) and the general regression neural network (GRNN), is proposed in this paper. Both FA and the GRNN are incremental learning systems and are very fast in network training. The proposed hybrid model, denoted as GRNNFA, is able to retain these advantages and, at the same time, to reduce the computational requirements in calculating and storing information of the kernels. A clustering version of the GRNN is designed with data compression by FA for noise removal. An adaptive gradient-based kernel width optimization algorithm has also been devised. Convergence of the gradient descent algorithm can be accelerated by the geometric incremental growth of the updating factor. A series of experiments with four benchmark datasets have been conducted to assess and compare effectiveness of GRNNFA with other approaches. The GRNNFA model is also employed in a novel application task for predicting the evacuation time of patrons at typical karaoke centers in Hong Kong in the event of fire. The results positively demonstrate the applicability of GRNNFA in noisy data regression problems.

In this paper, the effectiveness of three different operating strategies applied to the Fuzzy ARTMAP (FAM) neural network in pattern classification tasks is analyzed and compared. Three types of FAM, namely average FAM, voting FAM, and ordered FAM, are formed for experimentation. In average FAM, a pool of the FAM networks is trained using random sequences of input patterns, and the performance metrics from multiple networks are averaged. In voting FAM, predictions from a number of FAM networks are combined using the majority-voting scheme to reach a final output. In ordered FAM, a pre-processing procedure known as the ordering algorithm is employed to identify a fixed sequence of input patterns for training the FAM network. Three medical data sets are employed to evaluate the performances of these three types of FAM. The results are analyzed and compared with those from other learning systems. Bootstrapping has also been used to analyze and quantify the results statistically.

Applicability of an ensemble of Elman networks with boosting to drug dissolution profile predictions is investigated. Modifications of AdaBoost that enables its use in regression tasks are explained. Two real data sets comprising in vitro dissolution profiles of matrix-controlled-release theophylline pellets are employed to assess the effectiveness of the proposed system. Statistical evaluation and comparison of the results are performed. This work positively demonstrates the potentials of the proposed system for predicting desired drug dissolution characteristics in pharmaceutical product formulation tasks.

In this paper, an empirical study of the development and application of a committee of neural networks on online pattern classification tasks is presented. A multiple classifier framework is designed by adopting an Adaptive Resonance Theory-based (ART) autonomously learning neural network as the building block. A number of algorithms for combining outputs from multiple neural classifiers are considered, and two benchmark data sets have been used to evaluate the applicability of the proposed system. Different learning strategies coupling offline and online learning approaches, as well as different input pattern representation schemes, including the "ensemble" and "modular" methods, have been examined experimentally. Benefits and shortcomings of each approach are systematically analyzed and discussed. The results are comparable, and in some cases superior, with those from other classification algorithms. The experiments demonstrate the potentials of the proposed multiple neural network systems in offering an alternative to handle online pattern classification tasks in possibly nonstationary environments.

This paper investigates the efficacy and reliability of Artificial Neural Networks (ANNs) as an intelligent decision support tool for pharmaceutical product formulation. Two case studies have been employed to evaluate capabilities of the Multilayer Perceptron network in predicting drug dissolution/release profiles. Performances of the network were evaluated using similarity factor (f2) — an index recommended by the United States Food and Drug Administration for profile comparison in pharmaceutical research. In addition, the bootstrap method was applied to assess the network prediction reliability by estimating confidence intervals associated with the results. The Multilayer Perceptron network also demonstrated a superior performance in comparison with multiple regression models. The results reveal that the ANN system has potentials to be a decision support tool for profile prediction in pharmaceutical experimentation, and the bootstrap method could be used as a means to assess reliability of the network prediction.

The synergism of Web and phone technologies has led to the development of a new innovative voice Web network. The voice Web requires a voice recognition and authentication system incorporating a reliable speech recognition technique for secure information access on the Internet. In line with this requirement, we investigate the applicability of artificial neural networks to speech recognition. In our experiment, a total number of 200 vowel signals from individuals with different gender and race were recorded. The filtering process was performed using the wavelet approach to de-noise and compress the speech signals. An artificial neural network, specially the probabilistic neural network model, was then employed to recognize and classify vowel signals into their respective categories. A series of parameter settings for the PNN model was investigated and the results obtained were analyzed and discussed.

In this paper, a study of the effectiveness of a multiple classifier system (MCS) in a medical diagnostic task is described. A hybrid network, based on the integration of a fuzzy ARTMAP and the probabilistic neural network, is employed as the basis of the MCS. Outputs from multiple networks are combined using some decision combination method to reach a final prediction. By using a real medical database, a set of experiments has been conducted to evaluate the performance of the MSC with different network configurations. The experimental results reveal the potential of the MCS as a useful decision support tool in the medical field.