Goal-conditioned humanoid locomotion in reinforcement learning (RL) remains challenging due to sparse reward signals and the single-goal overfitting problem. Although contrastive reinforcement learning (CRL) has achieved considerable success in this setting, it can suffer from pronounced estimation variance, since epistemic uncertainty is difficult to reduce given limited task-specific information and model capacity. Ensemble-based critics can partially alleviate this issue. However, sufficient ensemble diversity and accurate individual estimates are not necessarily guaranteed during training, resulting in unstructured exploration. To address these challenges, we propose Conditional Redundancy-Guided Evolutionary Contrastive Ensemble with Direct Preference Optimization weighting (CRECE-DPO), which augments CRL with a vectorized critic ensemble and refines the ensemble via an evolutionary algorithm guided by a tailored fitness metric. Specifically, we design a DPO-weighted conditional redundancy fitness score, to prune redundant representations while promoting effective exploration of the parameter space. Simulation results on challenging goal-conditioned benchmarks, including humanoid locomotion, demonstrate consistent improvements over CRL and other baselines.

The existing camera tracking and perception methods mainly rely on sparse SLAM, which limits the dense perception ability of the scene and affects the reliability of auxiliary decision-making. Different from this, this work proposes a real-time tracking and unsupervised dense sensing framework. Firstly, the dense depth value of the scene is predicted by unsupervised multi-view stereo to remove the dependence on labeled data. Then, the quality of synthetic pseudo-reference image is quantified according to the predicted depth map and used as a weighted guidance to train the unsupervised model, thus reducing the ambiguity of feature matching in areas such as specular reflection. Moreover, the sparse optical flow of the keyframes is solved by real-time and robust ORB feature matching operator, which assists the high-precision training of unsupervised depth inference model. To increase the prediction accuracy of occluded area, a novel rendering consistency loss via neural radiance fields is designed to constrain the geometric characteristics of object surface. Finally, dense direct image alignment is performed from a global model to improve the tracking robustness, which is incrementally constructed from dense depth prediction. Extensive experiments on synthetic datasets and real datasets validate the effectiveness and practicability of the proposed work, which is an effective supplement to the existing SLAM work.

Series Snake robots possess strong adaptability for unstructured environments, but their trajectory tracking control is hindered by nonlinear dynamics and model uncertainties. This article proposes an optimal tracking control strategy based on an actor–critic reinforcement learning framework. The method integrates line-of-sight guidance with serpentine gait generation, while a neural network identification system approximates the solution of the Hamilton–Jacobi–Bellman equation for unknown dynamics. Actor and critic networks are employed to update control policies and cost functions online, reducing dependence on precise models. Rigorous theoretical analysis proves that position and velocity errors achieve semi-global uniform ultimate boundedness. Both simulations and prototype experiments were conducted based on a servo-driven yaw-pitch linkage alternating series snake robot. The results verify that the proposed method can achieve accurate trajectory tracking, rapid convergence, and stable joint control, demonstrating its effectiveness and superiority compared to existing methods.

Cognitive state detection holds significant research value in the field of human–computer interaction and neural engineering. However, existing works are insufficient in modeling the temporal dynamics of multimodal physiological signals, which leads to heterogeneous distribution differences in cross-modal feature interactions. In addition, domain shift issues under cross-subject and few-sample conditions restrict the model generalization performance. To cope with these problems, this work proposes a cognitive state detection framework that integrates Transformer-based multimodal feature interaction and self-training of pseudolabel optimization. First, the multihead attention mechanism is introduced to model the temporal evolution patterns across modalities, dynamically harmonizing cross-modal contributions to extract cognitive state-related shared features. Then, a dual-model cross-validation strategy is designed to filter high-quality pseudolabeled samples from the target domain for subsequent self-training, effectively avoiding the dependency on auxiliary modules in domain adaptation. Finally, Extensive experiments show that the proposed work significantly improves the recognition accuracy, and the designed pseudolabel optimization mechanism can be transferred to related tasks without increasing model complexity.

It requires terrain classification for unmanned Mars Rover to identify the safe areas. The current deep learning-based semantic segmentation and object recognition suffer from a large number of parameters and long training time. In this paper, a lightweight segmentation framework called Mobile-DeepRFB is proposed for the Martian terrain classification. It improves from the DeepLabV3$+$by taking the MobileNetV3 as the backbone module to decrease the parameters and the Receptive Field Block (RFB) module to strengthen the feature extraction capability as well as to enlarge the receptive field. Experimental results on the NASA Mars terrain dataset AI4MARS show that the presented method reduces 94% about the parameter number and improves the mean pixel accuracy by 2% compared to the existing ResNet101 and Xception backbone networks. The deployment of this framework on a low-computing power embedded platform (NVIDIA Jetson Xavier) demonstrates its great potential to apply to Mars rovers.Note to Practitioners—This paper was motivated by the problem of terrain classification of planetary rovers. Existing methods are typically based on semantic segmentation technology to recognize various terrains while suffering from the drawback of a large number of parameters. We propose a lightweight segmentation framework to address this issue. In particular, the lightweight backbone network is applied to significantly reduce the number of parameters. The receptive field module is substantially improved to enhance the feature extraction capability. Eventually, we deploy the framework on a low-computing platform. Experimental tests show that the framework can significantly reduce the number of network parameters and it can be used for planetary rovers with limited computational resources.

Given a set of calibrated images, Multiple View Stereo (MVS) applies end-to-end depth inference network to recover scene structure. However, previous methods designed pixel-visibility modules to aggregate cross-view cost, ignoring the consistency assumption of 2D contextual features in the 3D depth direction. The current multi-stage depth inference model also relies on intensive depth samples, which requires high memory consumption. To alleviate these problems, this work exploits edge-assisted epipolar Transformer for multi-view depth inference. The improvements of this work are summarized as follows: 1) The epipolar Transformer block is developed for reliable cross-view cost aggregation, and the edge detection branch is designed to constrain the consistency of epipolar geometry and edge features. 2) The dynamic depth range sampling mechanism based on probability volume is adopted to improve the accuracy of uncertain areas. Comprehensive comparisons with the state-of-the-art works indicate that our work can reconstruct dense scene representations with limited memory bottleblockNote to Practitioners—Learning-based MVS can obtain dense point clouds with accurate depth map estimation, which are widely applied in the fields of unmanned driving, battlefield environment perception and robot navigation. MVS-based scene reconstruction technology is the premise of the subsequent planning, decision-making and control of the human-machine system. To obtain dense scene representation with limited memory and runtime, this work proposes a multi-view stereo network with edge-assisted epipolar Transformer. Experiments on public benchmarks verify the feasibility and effectiveness of our model, which has good potential in battlefield environment reconstruction and human-computer interaction fields, and can provide intuitive and dense scene representation for decision-making assistance.

In lunar exploration missions, tasks such as target recognition or obstacle detection always encounter significant challenges due to the quality of images and the limited availability of training samples. Typically, existing methodologies can only recognize and locate a limited number of targets within simple scenarios, and models developed within simulated environments lack practical applicability. To alleviate above issues, an image fragments-based twofold learning strategy has been conceptualized. This strategy facilitates the decoupling of target features from environmental attributes through processes such as image fragmentation, confidence evaluation, and domain transfer, thereby achieving enhanced recognition accuracy across diverse lighting conditions. To provide more accurate localization and identification of a wider variety of targets,$\mathrm {\omega }$-CIoU loss is introduced to address the anomalies in prediction box scale and target count induced by flare and shadow features. Moreover, the establishment of the Real Chang’e Lunar Landscape Dataset, the most extensive public dataset for lunar surface target recognition in terms of sample quantity and diversity, provides an invaluable experimental foundation for future research in this field. Comprehensive experimental results on the RCLLD and public dataset ALLD demonstrate that the FTLS can significantly boosts recognition precision and generalization capabilities without escalating model complexity, outperforming prevailing target recognition methodologies for lunar exploration rovers.Note to Practitioners—This research is inspired by the need for high-precision, robust target recognition and avoidance systems for lunar rovers during lunar surface exploration missions. These systems are crucial for deployment on rovers with limited computational resources and energy, providing a convenient platform for future lunar terrain surveys, lunar base site selection, and mineral collection tasks. However, existing target recognition systems usually require training on extensive actual lunar datasets and often suffer from significantly reduced accuracy in areas with severe lighting changes, which decreases reliability on the lunar rover platform—an unacceptable risk for the cost-intensive lunar exploration missions. To address these challenges, this paper proposes a twofold learning strategy based on image fragments that trains on limited actual lunar data, offers resistance to flare and shadow interference, and its effectiveness is validated on both actual lunar and simulated datasets.

Due to model uncertainty, system coupling and external disturbance, the mathematical structure of control system is often constructed imprecisely. This unknown structure implies the unknown system dynamic and control input direction. Hence, it is difficult for unknown-structure Multiple Input Multiple Output (MIMO) coupled control systems to prescribe the settling time and control accuracy. To solve this problem, a novel prescribed-time robust direction-adjusting control method is proposed by adaptive Takagi-Sugeno (AT-S) fuzzy approximation. More specifically, by constructing an adaptive T-S fuzzy approximator under relaxed conditions, the mismatched uncertainty of systems can be transformed into the bounded matched uncertainty. It is a pre-condition for prescribed finite-time stability. After the prescribed time terminal, an adaptive robust control designed on AT-S fuzzy model is used for prescribed-accuracy convergence. Besides, compared with existing studies, the proposed method is independent of the structure information and initial control direction. Simulations and experiments verify its effectiveness. Note to Practitioners—The robot arm system with no or lowaccuracy Lagrange dynamic identification is a typical unknownstructure MIMO coupled system. It is difficult to achieve fastconvergence and high-accuracy control for this practical system, especially with no empirical pre-adjustment of the initial input direction. To solve this practical problem, a novel prescribedtime direction-adjusting adaptive T-S fuzzy control method is proposed. Firstly, the robotic arm system is decoupled into multiple joint subsystems by virtual dimension reduction strategy to avoid rule explosion, where system coupling is also converted to bounded unmatched uncertainty. Then, the inappropriate initial direction of the actuator input signal is addressed by control direction-adjusting algorithm. Meanwhile, based on adaptive TS fuzzy approximation theorem, the mismatched uncertainty is transformed into the bounded matched uncertainty. In the end, according to the prediction of approximation accuracy, a prescribed-accuracy robust controller is designed on AT-S fuzzy model. The proposed method is effective in a practical robotic arm control experiment without identification of Lagrange dynamics. Moreover, compared with other related methods, the control precision of the proposed method is higher.

Designing a robust air transportation network is critical for aviation activities to maintain highly efficient operations. Given certain total route cost, designing a strategy to improve the robustness of the network is a challenging issue. In this paper, first, a comparison experiment based on a small example shows the total effective resistance is a superior measure satisfying two critical criteria that other measures do not satisfy. Consequently, we explicitly formulate the network optimization problem as the minimization of total effective resistance under several constraints. The problem is often too time-consuming to solve to the optimal by exact methods. To achieve better efficiency, we propose a convex relaxation method for medium-scale networks making use of the problem properties. For large-scale networks, a clustering based convex relaxation method, reducing the dimensions of the network via selecting critical airports, is further proposed. To demonstrate the efficiency and efficacy of the proposed methods, simulations are performed on nine typical scale-free network and two real air transportation networks including Jetstar Asia Airway and domestic American Airlines.Note to Practitioners—The crucial need for designing robust air transportation networks in the aviation industry serves as the impetus for this work. The application scenarios range from the design of airport and flight operation networks for airlines, to air traffic networks with airways and waypoints, extending to drone cargo networks. An essential characteristic of a well-designed network is its ability to preserve connectivity, as much as possible, when it experiences disruptions such as flight cancellation or airway navigation equipment failure. The paper presents an argument for using total effective resistance as the key measure of network robustness, showing it to outperform other measures. We’ve developed specialized algorithms that leverage the unique properties of this optimization problem to minimize the total effective resistance of air transportation networks subject to certain total route cost to enhance the network structure. Although this study focuses on air transportation networks, the applicability of the algorithms extends beyond this field. Practitioners can apply them in other scenarios to design robust networks in various aviation operations.

In future battlefields, missiles are expected to become highly precise and efficient strike weapons, with missile intelligence emerging as a critical development trend. To address the problem of optimizing 3-D missile interception guidance laws, this article introduces the deep Q-network (DQN) algorithm on the foundation of proportional navigation guidance (PNG) and proposes an adaptive proportional guidance algorithm based on deep reinforcement learning (DRL). The proposed algorithm uses air combat situational information as the state space and incorporates parameters such as the missile-target relative distance and line-of-sight (LOS) angle into the reward function design. The optimal proportional navigation coefficient$K^{*}$for low-overload maneuvering targets is determined through network search, and the longitudinal and lateral control commands of the missile are decoupled by designing the proportional coefficient increment$\Delta K$, constructing a discretized action space. Simulation results show that, compared to the PNG with a constant$K^{*}$, the proposed method significantly improves the hit probability of high-overload maneuvering targets while maintaining the hit rate for low-overload maneuvering targets. As an exploration of future intelligent combat scenarios, this guidance law design method holds both theoretical significance and practical application value.

This work addresses the design problem of the fault detection observer (FDO) based on dynamic event-triggered mechanism for Markov jump systems under denial-of-service (DoS) attacks. The concept of limited energy for attackers is employed to characterize the property of nonperiodic DoS attacks. A dynamic event-triggered mechanism is introduced to save the system's communication resources. The $H_{\infty }/H_{-}$ index is incorporated to ensure that the designed FDO possesses both robustness against disturbances and sensitivity to faults. After obtaining a set of nonlinear inequalities using Lyapunov functional techniques, a simulated annealing algorithm is employed to assist in solving, ensuring not only the discovery of global optimization solutions but also obtaining satisfactory parameters for the dynamic event-triggered mechanism. Finally, the effectiveness of the designed FDO is illustrated by an example of a vertical take-off and landing vehicle dynamical system.

This article proposes a robotic fish formation control scheme that is based on finite-time terminal sliding mode to achieve coordinated tracking of multiple target paths under external disturbances and internal parameter perturbations. The method explores the lateral sliding mechanism of each body in the surge and sway directions by constructing a directional compensation guidance strategy that is based on a finite-time disturbance observer, thus enabling the precise coordinated movement of multiple robotic fish. Furthermore, this article acknowledges that the highly coupled dynamics of the robotic fish are susceptible to external environmental influences and modeling accuracy. Hence, a rapid global terminal sliding mode fuzzy controller that considers tangential displacement is introduced, and fuzzy adaptive methods are utilized to fit complex uncertainties. This approach mitigates the chattering problems commonly associated with traditional sliding mode control and enhances the error convergence speed and accuracy of the robotic fish formation system.

The industrial knowledge graph (IKG) can improve the cognitive intelligence of the manufacturing system and is recognized as one of the cores of the next-generation industrial management information system. Due to the multisource heterogeneous nature of industrial data, aligning entities with the same semantics (entity alignment) is the core technology for building large-scale, high-coverage IKGs. Existing approaches show that embedded learning of IKGs performs well for this task. However, most advanced methods ignore concept information when learning topological information about IKGs. Inspired by the ontology matching theory, in this article, we realize the importance of entity concepts in alignment. The conceptual semantics of entities can usually be obtained through the is–a relation. However, the IKG is usually constructed by triples (entity, relation, entity) automatically extracted from a large text corpus. This will lead to entities in the IKG having problems such as lacking conceptual information, belonging to multiple concepts, or having different concept granularities. To solve the two problems of lacking conceptual information and different concept granularity, we propose the concept-aware entity alignment network (CAEA), aggregating bidirectional relations and attributes to get the entity concept semantics by a novel concept-aware graph attention mechanism. The excellent performance of the CAEA can better support the construction of large and complete IKGs and support downstream applications such as industrial knowledge recommendation and assisted decision-making. To verify the performance of the CAEA on the IKG, we construct a new entity alignment benchmark using industrial control network security data and verify the effectiveness of the CAEA on the new benchmark and several mainstream datasets. Experimental results show that our method outperforms other state-of-the-art (SOTA) methods and promotes the development of IKGs.

Situation awareness (SA) is a crucial factor affecting flight safety for pilots, yet few studies have focused specifically on modeling SA for pilots, resulting in limited success. In this paper, we propose a novel multimodal deep learning approach to monitor pilots’ SA. The approach combines handcrafted and deep features obtained from eye movement and flight control data collected from 27 novice pilots across different training phases using a flight simulator. Ground truth SA measurements were obtained using the Situation Awareness Global Assessment Technique (SAGAT). The handcrafted features included 13 eye movements and 22 flight control features, while deep features were extracted from time-series of gaze positions using a deep extractor based on Transformer. By fusing the handcrafted features of eye movement and flight control, along with one deep feature of eye movement, we predicted the final SA level. Through leave-one-flight-out cross-validation, our model achieved a higher accuracy of 92.04%. The results indicate that the multimodal model outperforms the unimodal models, with the eye movement modality demonstrating superiority over the flight control modality in predicting SA. This suggests our method provides an objective means of predicting pilot’s SA and offers new insights for SA assessment in aviation and other fields. Overall, our multimodal deep learning approach holds promise for enhancing pilot training and flight safety by facilitating a more comprehensive understanding of pilots’ SA during critical flight scenarios.

This paper studies the problem of wide working conditions operational performance optimization of steam turbine units in coal-fired thermal power plants. A method based on big data association rule mining is proposed to establish relationships between a set of key controllable operating parameters and heat consumption rate of the units and then to determine optimal target values of the operating parameters. To deal with sparse data discretized from continuous unit operating data, a new association rule mining algorithm is proposed firstly. It utilizes a binary matrix to store data and introduces a search technique which combines linked list with pointers to mine frequent itemsets. These greatly eliminate the impact of sparsity of data i.e., large differences among transactions and scattered item distribution, on performance of the mining algorithm. Furthermore, in order to process large-scale data efficiently, a parallel implementation of the algorithm on Apache Spark platform is given. Finally, taking a steam turbine in a 600MW subcritical thermal power unit in China as an example, an overall optimization procedure including association rule mining and target value acquiring is presented. The mining results show that the target values calculated by the proposed optimization method can reduce heat consumption rate of the unit and improve economic benefits of the power plant significantly.Note to Practitioners—This research is motivated by the urgent requirement for energy saving and pollutant emission reducing in todays coal-fired power plants, especially in old thermal power plants. To achieve this objective, an effective method is to determine target values (also known as optimal values or benchmark values) of some key controllable operating parameters to realize operating optimization of the thermal power units. Existing target value determination approaches usually adopt the design values or the thermodynamic calculation values, which cannot truly represent the changes of unit operating characteristics. This research proposes a new big data-driven steam turbine unit target value determination method using association rule mining to reduce heat consumption rate of the unit. We not only design a new association rule mining algorithm but also give a parallel implementation of it on advanced big data distributed parallel computing framework Apache Spark, making the proposed optimization method efficiently mine large scale operating data. The proposed operating optimization method is based on the actual operating data of the unit. It has high reliability and good general applicability for different types of coal-fired power units. The optimization experiment on a steam turbine unit in China shows that the obtained target values are valuable for guiding optimization operation of the unit under wide working conditions. In the future, we will consider explicit pollutant emission in the optimization index and address multi-index operating optimization of the steam turbine unit.

Chronic disease patients often require revisits for long-term care. Online medical services shift revisits to online, which can improve the access to chronic care and reduce the burden on offline medical services. However, whether Internet healthcare can truly match the medical supply and demand, one of the critical issues is the efficient advance scheduling of the integrated online and offline systems. This study investigates the advance scheduling problem for the first visit and revisit patients in chronic care. The uncertainty of revisit status (i.e., online or offline) and heterogeneity of online and offline revisits (i.e., revisit interval, continuity of care violation penalty) are considered. A stochastic mixed-integer programming model is formulated for assigning patients to a specific physician on a specific day over the course of a finite planning period. The aim is to minimize the expected sum of three cost components related to offline and online services: overtime and idle time, continuity of care violation penalty, and fixed setup. This study proposes a modified progressive hedging algorithm and applies a sequential decision-making framework to obtain rolling time advance schedules. Results of the numerical analysis demonstrate the effectiveness of our algorithm compared to both the published state-of-the-art Lagrangian decomposition embedded with surrogate subgradient method and the commercial solver Gurobi. The insight obtained from the experiments is that a capacity allocation scheme with all physicians assigned with both offline and online capacities would be a good choice for considerable cost savings.Note to Practitioners—Internet healthcare is becoming increasingly popular. Operation and management issues have arisen in the integrated online and offline appointment systems. A sequential decision-making method embedded with a stochastic programming model and a modified PHA is proposed to help decision-makers generate the first visit and revisit advance schedules for chronic care. The performance of this approach and the system is thoroughly verified. Results show that the developed decision technique can lessen the operational cost generated by scheduling and realize the goal of continuity of care. This study offers a useful tool to help with intelligent patient advance scheduling in an integrated management system of online and offline chronic care.

Time granularity is a key parameter necessary for short-time passenger flow prediction of urban rail transit (URT); however, no universal method is available for its setting. This study investigates the time granularity setting principle for short-term passenger flow prediction in URT. First, a method to measure the autocorrelation of passenger flow time series is constructed, focusing on the comparison of time granularities. Second, based on the functional relationship between the first-order autocorrelation coefficients of the passenger flow time series under different time granularities, the time granularity setup principle is obtained for different passenger flow characteristics. Finally, the reasonableness and universality of the time granularity setting principle are verified by analyzing the passenger flow characteristics and autocorrelation magnitude of the actual inbound and origin-destination (OD) passenger flow data under different stations and dates at different time granularities.

The similarity measurement method is the key part of the normal cloud model as well as its applications. Too much attention has been paid on geometric and numerical features, leading to the weak interpretability and unreasonable results in the similarity measurement method of the normal cloud models. A bidirectional and weighted similarity measurement (BWSM) method is proposed by the number distribution and membership of cloud droplets on a normal cloud model. First, the cloud droplets are analyzed, and their characteristics of number distribution and membership that belong to different cloud models are obtained. Second, the similarity measurement strategy of the normal cloud models is analyzed and used to propose a BWSM method, which is compared with two commonly used similarity measurement methods of normal cloud models. The influence of expectation, entropy, and hyper entropy on the similarity measurement method of the normal cloud models is analyzed. Finally, the method is applied on the operational status perception and computing of urban rail transit. Results show the method has high rationality, validity, and applicability for perceiving and computing the operational status of urban rail transit. The method improves the interpretability of the similarity measurement of the normal cloud models and the credibility of the research based on the similarity measurement of the normal cloud models.

The emergency plan of the urban rail transit (URT) system is a guiding document for dealing with emergencies and formulating emergency plans. However, the emergency plan text described in natural language has some problems, such as poor visibility and enforceability. Effective help is difficult to provide for the rapid implementation of emergency response. Therefore, obtaining key emergency task information from the emergency response plan and visualizing the emergency disposal workflow are the main challenges. In this article, we propose a method of extracting emergency elements (EELs) from emergency plans and constructing a business process model. First, a nested entity extraction model incorporating adversarial training is proposed to extract EELs from the complex sentences in the emergency plan text. Second, the EELs are combined into emergency task units, and then the relations between emergency task units are identified to form the emergency task sequence flow, which is stored in matrix form. Finally, the emergency disposal workflow model is generated based on the emergency task sequence flow and the BPMN modeling method. Taking the actual emergency plan text as an example, the process from the extraction of EELs to the construction of the disposal workflow model is demonstrated. Experimental results prove that this method has advantages in comprehensively extracting EELs, visualizing the emergency disposal workflow, and improving the enforceability of emergency plans.

The application of reinforcement learning (RL) in artificial intelligence has become increasingly widespread. However, its drawbacks are also apparent, as it requires a large number of samples for support, making the enhancement of sample efficiency a research focus. To address this issue, we propose a novel N-step method. This method extends the horizon of the agent, enabling it to acquire more long-term effective information, thus resolving the issue of data inefficiency in RL. Additionally, this N-step method can reduce the estimation variance of Q-function, which is one of the factors contributing to estimation errors in Q-function estimation. Apart from high variance, estimation bias in Q-function estimation is another factor leading to estimation errors. To mitigate the estimation bias of Q-function, we design a regularization method based on the V-function, which has been underexplored. The combination of these two methods perfectly addresses the problems of low sample efficiency and inaccurate Q-function estimation in RL. Finally, extensive experiments conducted in discrete and continuous action spaces demonstrate that the proposed novel N-step method, when combined with classical deep Q-network, deep deterministic policy gradient, and TD3 algorithms, is effective, consistently outperforming the classical algorithms.

This work presents a butterfly spiral propulsion mode of snake robots to realize the tracking control on the objective trajectory in multiple scenarios. This method investigates the force mechanism of each body element in the yaw and pitch directions. The butterfly spiral gait mechanic and friction models are constructed to offset the lateral torque force caused by joint rotation. In addition, this work combines an integral part of improving the line of sight guidance scheme, which eliminates the robot's sideslip when tracking the curve track and enhances the body's adaptability to different scenarios. Lyapunov's theory proves the stability of the designed guidance strategy. Simulation and experimental results illustrate that the designed butterfly spiral gait and guidance scheme can provide the snake robot faster tracking results and more stable error performance than the cylindrical and conical spiral gait.

Given the precalibrated image pairs, stereo matching aims to infer the scene depth information in real-time, which has important research value in the fields of high-precision 3-D reconstruction of the Earth’s surface, automatic driving and unmanned aerial vehicle (UAV) navigation. The cost volume-based stereo matching method adopts a coarse-to-fine manner to construct cascaded cost volume, and applies 3-D convolution to capture the correspondence of feature matching to infer the disparity map, which achieves comparable performance. However, the existing method has difficulty dealing with jitter regions with disparity change, and direct disparity regression easily leads to overfitting of cost volume regularization. To alleviate the above two problems, this work proposes an end-to-end disparity estimation network based on Transformer. Its specific improvements are as follows. 1) The cross-view feature interaction module based on Transformer is introduced to realize the feature interaction of global context information. 2) A parallax attention mechanism is designed to impose global geometric constraints on the epipolar line to improve the reliability of feature matching. 3) Focal loss is applied for the training of the disparity classification model to emphasize one-hot supervision in ambiguous regions. Comprehensive experiments on public datasets Sceneflow, KITTI2015, ETH3D, and aerial WHU datasets validate that the proposed work can effectively enhance the performance of disparity estimation.

Urban rail transit (URT) is vulnerable to natural disasters and social emergencies including fire, storm and epidemic (such as COVID-19), and real-time origin-destination (OD) flow prediction provides URT operators with important information to ensure the safety of URT system. However, hindered by the high dimensionality of OD flow and the lack of supportive information reflecting the real-time passenger flow changes, study in this area is at the beginning stage. A novel model consisting of two stages is proposed for OD flow prediction. The first stage predicts the inflows of all stations by Long Short-Term Memory (LSTM) in real time, where the dimension is reduced compared with predicting OD flows directly. In the second stage, the notion of separation rate, namely, the proportion of inbound passengers bounding for another station, is estimated. Finally, The OD flow is predicted by multiplying the inflow and separation rate. Experiments based on Hangzhou Metro dataset show the proposed model outperforms the contrast model in weighted mean average error (WMAE) and weighted mean square error (WMSE). Results also suggest that the proposed prediction model performs better on weekdays than on weekends, and with greater accuracy on larger OD flows.

This work provides a graph network solution for pilot brain fatigue state inference based on electroencephalography (EEG) fatigue indicators. Two graph methods are built as follows. The first one uses a single EEG signal sample as a node, and fatigue detection as a node classification task in a graph network. The developed graph network is then utilized to extract the correlation among different samples to achieve multisample joint decision making. The second method uses a single EEG signal sample as a graph structure, and EEG fatigue prediction as a graph classification task. Electrode position correlation is used to construct a graph. The feature fusion of adjacent electrodes is obtained through the connection relationship among nodes in a graph structure to improve network learning accuracy. In addition, a Bayesian optimization method is proposed to model the randomness of attention weights, and a Bayesian graph attention network is built. This work constructs a based-graph deep learning structures to achieve a pilot fatigue detection model with high accuracy, good generalization, and strong adaptability. Experimental results demonstrate the effectiveness of the proposed model.

Multiagent cooperative systems can be used to conceptualize many real-world problems. Reinforcement learning is a particularly effective tool. The issue of bias in$Q$-function value estimation in single-agent reinforcement learning has garnered a lot of interest and substantial study. Indeed, this challenge endures in multiagent reinforcement learning, primarily owing to the inclusion of maximization operations. The crux of the matter lies in the inability to seamlessly extrapolate single-agent reinforcement learning algorithms to their multiagent counterparts. In this article, we introduce a more encompassing and straightforward principle: the notion of appropriate value correction. We suggest replacing the maximization operation with a monotonically nondecreasing function to obtain more accurate value estimates. We theoretically demonstrate that this operation effectively reduces the potential overestimation bias in the QMIX algorithm. Ultimately, our methodology, dubbed the SMIX algorithm—a fusion of the QMIX algorithm empowered by the Softmax operator, attains state-of-the-art outcomes across diverse multiagent cooperative tasks. This success extends to challenging domains such as StarCraft II, marking it as one of the most formidable games to date.

Multiple-view stereo has potential applications in robotic operations and autonomous driving (unstructured environment construction, visual servo). With assisted depth information, inertial navigation systems can achieve precise navigation. It is, especially suitable for GPS failures in complex environments. Accurate depth estimation is a challenge in low-textured or occluded regions. To alleviate the inference of incorrect depth, a multi-stage pixel-visibility learning-based stereo network is presented in this paper. Its improvements are as follows: 1) a new content-adaptive cost volume aggregation mechanism based on neighboring pixel-wise visibility is designed to effectively produce more accurate and smoother depth map predictions in the object boundary. 2) global convolution block and boundary refinement block are developed to regularize its cost volume, they can learn the inherent constraints of feature matching correspondence and effectively mitigate the depth estimation uncertainty in low-textured regions. 3) a new loss function is designed to measure the uncertainty of predicted probability distribution and enhance the reliability of depth map inference. Experimental results on the indoor DTU datasets and the outdoor Tanks & Temples datasets indicate that our method can achieve superior performance and has a powerful generalization ability, which is comparable to state-of-the-art works. Note to Practitioners—Multiple-view stereo (MVS) can estimate dense 3D representations of scenes, which is widely used in autonomous driving, robotic navigation, virtual reality (VR), and augmented reality (AR). Aiming at the problem of incorrect depth inference in low-textured or occluded regions, this work proposes a novel multi-stage depth prediction method based on neighboring pixel-wise visibility. Our method cannot only achieve accurate depth estimation for robot perception but also make no concession to real-time performance. It is clear that the proposed method has good potential in 3D reconstruction, robotic navigation, and VR/AR fields to provide accurate depth estimation in real-time with limited memory consumption.

Semi-supervised learning has attracted wide attention from many researchers since its ability to utilize a few data with labels and relatively more data without labels to learn information. Some existing semi-supervised methods for medical image segmentation enforce the regularization of training by implicitly perturbing data or networks to perform the consistency. Most consistency regularization methods focus on data level or network structure level, and rarely of them focus on the task level. It may not directly lead to an improvement in task accuracy. To overcome the problem, this work proposes a semi-supervised dual-task consistent joint learning framework with task-level regularization for 3D medical image segmentation. Two branches are utilized to simultaneously predict the segmented and signed distance maps, and they can learn useful information from each other by constructing a consistency loss function between the two tasks. The segmentation branch learns rich information from both labeled and unlabeled data to strengthen the constraints on the geometric structure of the target. Experimental results on two benchmark datasets show that the proposed method can achieve better performance compared with other state-of-the-art works. It illustrates our method improves segmentation performance by utilizing unlabeled data and consistent regularization.

Recognition of amount in figures in the financial multi-bill scenes is crucial for the automatic banking business. However, the diversity of banking business and the limitation of customer data privacy determine that it is difficult to collect a large number of sample datasets. Aiming at the problem of insufficient training data in multi-bill scenes and the low accuracy of the detection model, this article proposes a new generative adversarial network (GAN) to generate new samples and to expand the bill dataset, which is then adopted to train a framework for recognition of the bill amount. In the proposed WGAN-SA, a residual block is adopted as the basic structure of the generator and the discriminator, and the self-attention mechanism is also utilized to improve the generation performance. In addition, Wasserstein distance is utilized to measure the distance between real and synthetic samples. Experimental results on the benchmark dataset and comparisons with state-of-the-art works show that our proposed WGAN-SA can effectively improve the few-sample learning performance. Besides, experiments on the bill dataset verify that our method can solve the problem of model collapse and has the ability to generate images of the amount in figures with better fidelity and variety, which is also helpful to achieve better bill amount recognition performance compared with other latest works.

Three-dimensional spiral gait enables a snake robot to climb over obstacles, cross caves, and adapt to complex environments. This article reports an antidisturbance path-following control method for a snake robot with a spiral gait. This method reduces the deviation of the robot's position in following the ideal path by estimating the time-varying parameters, the external disturbances, and the viscous friction coefficients. The estimations are used to compensate for the control inputs of the system, which can improve the adaptability of the robot to the environment. Then, the attitude and position errors can rapidly converge to the origin. An appropriate Lyapunov function is adopted to explore the stability of following errors. Experimental results show that the proposed method can accelerate the convergence rate of errors, reduce the fluctuation peak, and improve the following stability of snake robots.

Redundant robot manipulators play a significant role in modern industry. In this article, we propose a solution scheme to the trajectory tracking problem of the redundant robot manipulator with physical constraints through the Zhang neural dynamics method. Such problem is integrated into a time-varying system consisting of time-varying nonlinear equation (TVNE) and time-varying linear inequality (TVLI) and solved online by the varying-parameter Zhang neural dynamics (VPZND) model. It is ensured that the redundant robot manipulator can still perform the tracking task perfectly under the coexistence of time-varying bounded noise and physical constraints. Theoretical analysis proves that this VPZND model also has an explicit fixed convergence time. Numerical experiments confirm the feasibility of our VPZND model for TVLI. The trajectory tracking problem of the redundant robot manipulator with six or three degrees of freedom under the dual influence of physical constraints and noise is perfectly solved by the VPZND model, which is enough to verify its practical value.

In the context of Industry 4.0, the medical industry is horizontally integrating the medical resources of the entire industry through the Internet of Things (IoT) and digital interconnection technologies. Speeding up the establishment of the public retrieval database of diagnosis-related historical data is a common call for the entire industry. Among them, the Magnetic Resonance Imaging (MRI) retrieval system, which is one of the key tools for secure and private the Internet of Medical Things (IoMT), is significant for patients to check their conditions and doctors to make clinical diagnoses securely and privately. Hence, this paper proposes a framework named MRCG that integrates Convolutional Neural Network (CNN) and Graph Neural Network (GNN) by incorporating the relationship between multiple gallery images in the graph structure. First, we adopt a Vgg16-based triplet network jointly trained for similarity learning and classification task. Next, a graph is constructed from the extracted features of triplet CNN where each node feature encodes a query-gallery image pair. The edge weight between nodes represents the similarity between two gallery images. Finally, a GNN with skip connections is adopted to learn on the constructed graph and predict the similarity score of each query-gallery image pair. Besides, Focal loss is also adopted while training GNN to tackle the class imbalance of the nodes. Experimental results on some benchmark datasets, including the CE-MRI dataset and a public MRI dataset from the Kaggle platform, show that the proposed MRCG can achieve 88.64% mAP and 86.59% mAP, respectively. Compared with some other state-of-the-art models, the MRCG can also outperform all the baseline models.

Unmanned aerial vehicle (UAV)-assisted communication is a significant technology in 6G communication. In order to cope with the dynamic trajectory optimization problem of the air-ground network, the interaction between entities is modeled as a Markov game firstly. Then, the model-free multi-agent reinforcement learning (MARL) is adopted to optimize individual decision-making. This enables agents to learn the mobile patterns of others, so as to optimize their own mobile strategy. However, there are some common issues when executing the benchmark MARL algorithms, such as biased estimation and local optimum. To solve these problems, an enhanced multi-agent proximal policy optimization algorithm is proposed with policy clipping and average evaluation to guarantee the fast convergence and accurate estimation. Simulations demonstrate that this method produces superior convergence than the benchmark algorithms. It allows the UAV base station, ground users and the aerial jammer to adopt the optimal mobile strategies to achieve their respective maximum cumulative rewards. In addition, the stable strategies of agents constitute the approximate Nash equilibrium for the UAV-assisted communication Markov Game.

Flight Trajectory Prediction (TP) is an essential task in Air Traffic Control (ATC). Currently, the TP task is usually achieved by regression approaches, which concatenates several scalar attributes of the observation into a low-dimensional vector as the inputs. However, it is difficult to accurately model aircraft motion patterns using low-dimensional features in complex and time-varying ATC environments. To improve the performance of the TP task, in this paper, a novel framework, called FlightBERT, is proposed based on Binary Encoding (BE) representation, which enables us to tackle the TP task as a multi binary classification problem. Specifically, the scalar attributes of the flight trajectory are encoded into binary codes and transformed into a high-dimensional representation by the attribute embedding module. Considering the prior knowledge among flight attributes, an Attribute Correlation Attention (ACoAtt) block is designed to explicitly capture the correlations among the specific attributes. A stacked Transformer block is applied to serve as the backbone network, which is followed by the predictor to generate the outputs. Considering the nature of flight trajectory, a hybrid constrained loss, i.e., combining the mean square error loss with the binary cross-entropy loss, is innovatively designed to optimize the proposed framework. The proposed method is validated on a large-scale dataset, which is collected from the real-world ATC environment. The experimental results demonstrate that the proposed method outperforms other baselines by quantitative and qualitative evaluations.

To eliminate the influence of nonlinear state terms in the highly-coupled unmanned aerial vehicle (UAV) model and improve the aircraft’s ability to suppress wind field interferences, this work presents a path-following scheme for UAVs. This method uses the radial basis neural network (RBNN) to develop an adaptive approximation law for the gyroscopic effect function to balance for the influence of system uncertainty and nonlinear state terms on UAV modeling and reduce the dependence of the UAV’s roll and pitch control orders on attitude velocity information. In addition, the adaptive update laws of the disturbance predictions are designed to compensate for the control input and repress the chattering and deviation of the drone. The stability of the proposed controller was proven by using the Lyapunov theorem. Simulations and experiments have shown that the controller can perform faster convergence speed and higher following accuracy of the flight position and attitude errors.

In this paper, a graph neural network anomaly detection framework is proposed to improve the safety of linear induction motors, a key component of high-speed maglev trains. In our framework, each sensor sequence is treated as a separate feature. The similarity and correlation between multi-dimensional features are learned as prior knowledge for graph structure learning. The spatial-temporal graph attention network incorporates prior knowledge to learn complex correlations between nodes. Furthermore, the framework optimises a joint model for anomaly detection, avoiding the trap of falling into either local or global optimisation and thus achieving the most stable detection. Experimental results of our method on four real-world datasets show that it is more accurate than other state-of-the-art methods in detecting anomalies and capturing inter-sensor correlations. Further analysis of graph attention weights and visualization subgraphs show that our framework is well interpretable and allowing users to locate the root cause of anomalies.

How to efficiently collect multi-dimensional information of traffic accidents with high dangers has triggered intensive concerns in intelligent transportation community. This paper provides a viable and continuous observation solution using low-budgeted mobile robots. At the kinematic level, a relative position estimator utilizing available distance data is constructed with an exponential error convergence, removing the dependency of using global position. To proceed, resorting to estimation results, an elliptical encircling guidance rule with a time-varying radius is established to preserve circumnavigation concerning targets. At the kinetic level, a new fuel-saving uncertainty mitigation scheme, i.e., event-triggered unknown system dynamics estimators (ETUSDEs) with the feature of a reduced transmission load and a concise structure are respectively developed in velocity and angular rate subsystems to reconstruct uncertainties with a prescribed decaying rate, where event-triggering conditions are enforced to schedule updating frequency of actuators and measurements in an aperiodic manner instead of a fixed time interval. Then, an event-triggered robust kinetic control protocol is synthesized to achieve an accurate command tracking without incurring Zeno behaviors. Finally, the convergence of entire system is illustrated through input-to-state stable (ISS) criterion. Simulation results are delivered to testify the effectiveness of proposed method.

To respond to the rapid growth of shipping container throughput, terminals urgently need to improve the efficiency of thier operations and reduce operational costs through automation and intellectualization upgrades, thereby improving service levels and enhancing market competitiveness. Due to the advantages of reliable transportation, efficient operation, and environmental friendliness, AGV-based automated container terminal (ACT) has become the development trend of container terminals. To help ACT improve its operational management capabilities, plenty of scholars have explored the transportation system of ACT. Through the analysis of operational management issues, the paper defines the four main research topics in vehicle transportation of the ACT including equipment scheduling, path planning, exception handling, and vehicle management. Then, in each topic, the works in the recent 25 years are summarized and several research opportunities for possible follow-up research directions in different fields are proposed. We expect our survey could not only provide references for more scholars on the research of operation and management of terminals, but also provide guidance for system evaluation and improvement for terminal system engineers and operation managers.

In the context of establishing emission control areas (ECAs) in many ports to meet the challenges posed by air pollution, the use of drone-carrying sniffers to perform emission monitoring missions has become a new monitoring mode for ECAs. The operational management problem of drones in ECAs, namely, drone scheduling problem (DSP), is eliciting the attention of researchers. To consider the influence of vessel traffic on the demand for drones, this study proposes a rental-based drone operation model. In the model, the number of drones used depends on the load of monitoring missions. Maximizing the cumulative monitoring reward and minimizing the use number of drones within the minimum monitoring rate constraint are used as optimization objectives to maximize the cost return of the rental-based operation model. The rental-based drone operation model is modeled as a multi-objective DSP (MDSP). Furthermore, we horizontally compare the characteristics of MDSP with those of many classical models in the field of operations research. Afterward, we reveal the similarities and differences between MDSP and previous models. We find that MDSP has the non-first-in-first-out property, whereas most of the advanced models have the first-in-first-out property, which leads to the failure of the developed efficient algorithms in solving MDSP. Therefore, this study innovatively designs four feasible multi-objective optimization methods for MDSP. Numerical experiments are conducted to evaluate the performance of the four methods in solving MDSP with different scales. In terms of theoretical implications, experimental results prove that the proposed methods for solving MDSP are feasible and effective. In terms of practical implications, the proposed rental-based vessel monitoring operation model shows great potential for practical engineering.

This work reports an adaptive path-following controller for a multijoint snake robot (MSR) to improve the adaptability of the robot to the environment. The new strategy estimates the time-varying parameters of the system and the external interference to adjust the motion state of the robot in real time. Estimations are used to compensate for the joint torque of an MSR, thus reducing the fluctuation peak of path-following errors. In addition, this work designs an anti-sideslip line-of-sight (LOS) guidance strategy to avoid the deviation of the direction angle. The method can improve the tracking accuracy of an MSR, and the position errors enable the system to achieve uniformly ultimate boundedness (UUB). The angle errors converge to the origin to achieve stability. Experimental results demonstrate that the novel method can accurately estimate the time-dependent parameters, sideslip, and interference, raise the convergent speed of errors, and reduce the fluctuation peak.

This article proposes a batch-mode active learning (AL) method of Gaussian process regression (GPR), which is based on the expected model change maximization. Unlike existing strategies for measuring the model change, weight’s information gain (WIG) caused by model training is introduced to measure the model change in this article, which has a lower computational cost. First, two methods for calculating the WIG of the GPR are given from the perspective of accuracy and easy realization, respectively. Then, two AL algorithms are designed by using enumeration and greedy sampling. Finally, experiments on three datasets from various domains have verified the effectiveness of the proposed AL algorithms.

This work proposes a nonparametric prior induced deep sum-logarithmic-multinomial mixture (DSLMM) model to detect pilots' cognitive states through the developed brain power map. DSLMM uses multinormal distribution to infer the latent variable of each neuron in the first layer of the network. These latent variables obeyed a sum-logarithmic distribution that is backpropagated to its observation vector and the number of neurons in the next layer. Multinormal distribution is used to segment the extended observation vector to form a matrix associated with the width of the next layer. This work also proposes an adaptive topic-layer stochastic gradient Riemann (ATL-SGR) Markov chain Monte Carlo (MCMC) inference method to learn its global parameters without heuristic assumptions. The experimental results indicate that DSLMM can extract more probability distribution contained in the brain power map layer by layer, and achieve higher pilot cognition detection accuracy.

This work proposes a scalable gamma non-negative matrix network (SGNMN), which uses a Poisson randomized Gamma factor analysis to obtain the neurons of the first layer of a network. These neurons obey Gamma distribution whose shape parameter infers the neurons of the next layer of the network and their related weights. Upsampling the connection weights follows a Dirichlet distribution. Downsampling hidden units obey Gamma distribution. This work performs up-down sampling on each layer to learn the parameters of SGNMN. Experimental results indicate that the width and depth of SGNMN are closely related, and a reasonable network structure for accurately detecting brain fatigue through functional near-infrared spectroscopy can be obtained by considering network width, depth, and parameters.

Current brain cognitive models are insufficient in handling outliers and dynamics of electroencephalogram (EEG) signals. This article presents a novel self-paced dynamic infinite mixture model to infer the dynamics of EEG fatigue signals. The instantaneous spectrum features provided by ensemble wavelet transform and Hilbert transform are extracted to form four fatigue indicators. The covariance of log likelihood of the complete data is proposed to accurately identify similar components and dynamics of the developed mixture model. Compared with its seven peers, the proposed model shows better performance in automatically identifying a pilot's brain workload.

The drone has become a promising tool to improve the efficiency of vessel emission monitoring in emission control areas of the part due to its high mobility. However, how to optimize the flight path of drones to improve the weighted sum of monitored vessels, i.e., drone scheduling problem (DSP), is a not yet fully researched problem. In this paper, different from the classic optimization solution method used by the literature, an efficient ant colony-based algorithm is developed to solve DSP. Given the characteristics of DSP, a hierarchical-based pheromone update strategy and partition-based pheromone management mechanism are proposed to optimize the typical ant colony algorithm. Numerical experiments not only illustrate the feasibility of using the ant colony algorithm to solve DSP, but also show that the algorithm we proposed outperforms other compared methods in terms of the solution quality and the solving speed under different problem scales.

How to effectively organize drones to monitor pollutants from vessels is an important operational problem in port management. It is defined as the drone scheduling problem (DSP). The effectiveness of precise algorithms and heuristic algorithms in solving DSP has been reported in previous studies. In previous studies, the speed of the vessel was assumed to be constant. However, since the influence of sea waves and vessel power, such an assumption is difficult to satisfy in actual scenarios. The actual position of the vessel may deviate from the position information obtained through prior calculations. As the cumulative position deviation increases, it is possible to make the original feasible monitoring scheme infeasible. It is necessary to consider the emission monitoring dispatching of drones under vessel speed fluctuation in actual monitoring activities of the vessel. To deal with the problem, a dynamic dispatching strategy based on reinforcement learning (RL) is proposed. Considering the vessel speed fluctuation, the monitoring window is divided into multiple sub-time windows. The route information of the vessel in each sub-time window is updated according to the vessel speed fluctuations to reduce the accumulation of deviations between the prior position and the actual position. Then, a lightweight RL strategy is adopted to quickly (re)organize the monitoring scheme in each sub-time window. Numerical experiments illustrate the above division-conquer approach could effectively reduce the possibility of drone monitoring failure caused by vessel speed fluctuations. Also, the superiority of the RL-based dispatching strategy is illustrated by comparing it with multiple dispatching schemes.

Multiple-View Stereo (MVS) aims to reconstruct the dense 3D representations of scenes. MVS has potential applications in the fields of autonomous driving (unstructured environment construction) and robotic navigation (visual-inertial navigation). To mitigate the error of depth estimation in low-textured or occluded regions, this work proposes a two-stage multi-view stereo network for fast and accurate depth estimation. The improvements of this work over the state of the art are as follows: 1) Sparse costs are constructed to jointly predict the initial depth map and surface normal by cost regularization, which proves that the surface normals can be estimated in this way with low memory consumption. 2) A new edge refinement block is developed to refine the coarse surface normal to obtain a fine-grained surface normal map. 3) Instead of using the general variance-based metric to equally aggregate cost, a new content-adaptive cost aggregation mechanism based on the similarity of the neighboring surface normal is designed for reliable cost aggregation. To the best of our knowledge, the proposed work is the first trainable network that leverages surface normal as guidance to capture neighboring pixel-visibility, which is an effective supplement to existing depth/normal estimation frameworks. Experimental results indicate that our method can not only achieve accurate depth estimation for scene perception but also make no concession to the real-time performance and limited memory bottleblock. Multiple-view stereo (MVS) aims to reconstruct the dense 3D representations of scenes. It is widely used in the fields of industrial measurement, autonomous driving, and robotic navigation. To mitigate the error of depth estimation in challenging scenarios, this work proposes a two-stage multi-view stereo network for fast and accurate depth estimation. Our method is the first trainable network that leverages surface normal as pixel-visibility guidance to aggregate reliable cost, which could achieve accurate depth estimation and provide the perception ability for the robot. The proposed method has great potential in the fields of 3D reconstruction, industrial measurement, and robotic navigation to estimate real-time and accurate depth with limited memory consumption.

This work designs an adversarial Bayesian deep network to solve the cognitive detection of pilot fatigue. Batch normalization and data enhancement are adopted in the posterior inference of the proposed model parameters to effectively improve the generalization of neural networks. The generator is used to enhance the brain power map generated from three cognitive indicators and improve the accuracy of fatigue state recognition. This work also adds adversarial noise in the vicinity of each brain electrode to form an adversarial image, which further reveals the correlation between the cognitive state of brain and the location of brain regions. Compared with other deep models and parameter optimization methods, our model achieves better detection accuracy.

The evaluation of pilot brain activity is very important for flight safety. This study proposes a Hidden semi-Markov Model with Hierarchical prior to detect brain activity under different flight tasks. A dynamic student mixture model is proposed to detect the outlier of emission probability of HSMM. Instantaneous spectrum features are also extracted from EEG signals. Compared with other latent variable models, the proposed model shows excellent performance for the automatic inference of brain cognitive activity of pilots. The results indicate that the consideration of hierarchical model and the emission probability with${t}$mixture model improves the recognition performance for Pilots’ fatigue cognitive level.

The relationship between flight performance and multi-physiological parameters under different flight operating patterns is unknown. This work proposes a Stacked Gaussian Process Network (SGPN) to reveal it. SGPN is a multi-layer network model formed by recursion from a regular Gaussian process and random disturbance. This work constructs an auxiliary variable strategy with the induced points to improve its learning efficiency, thus leading to a sparse SGPN model. In it, a Gaussian process acts as an activation function of each node, but the entire model is no longer a Gaussian process and thus very challenging to solve it. This work presents its solution via variational approximate inference. Experimental results of pilot flight performance evaluation show that the proposed model has stronger learning and generalization ability than its seven competitive peers. It is able to approximate non-linear coupling relationship between multi-physiological parameters and flight height differences.

This article proposes a four-dimensional brain mapping method, which can represent the continuous process of a person’s fatigue state in the form of image frames in a space-time range. This work couples 3-D-convolutional neural networks and long-short-term memory networks to form a cognitive detection model of brain fatigue dynamics, which can simulate the continuous process of a person’s brain fatigue dynamics and accurately identify different cognitive fatigue states. Our approach can be applied to any type of brain fatigue detection.

Pilots' brain fatigue status recognition faces two important issues. They are how to extract brain cognitive features and how to identify these fatigue characteristics. In this article, a gamma deep belief network is proposed to extract multilayer deep representations of high-dimensional cognitive data. The Dirichlet distributed connection weight vector is upsampled layer by layer in each iteration, and then the hidden units of the gamma distribution are downsampled. An effective upper and lower Gibbs sampler is formed to realize the automatic reasoning of the network structure. In order to extract the 3-D instantaneous time-frequency distribution spectrum of electroencephalogram (EEG) signals and avoid signal modal aliasing, this article also proposes a smoothed pseudo affine Wigner-Ville distribution method. Finally, experimental results show that our model achieves satisfactory results in terms of both recognition accuracy and stability.

How to track the attention of the pilot is a huge challenge. We are able to capture the pupil status of the pilot and analyze their anomalies and judge the attention of the pilot. This paper proposes a new approach to solve this problem through the integration of spherical Haar wavelet transform and deep learning methods. First, considering the application limitations of Haar wavelet and other wavelets in spherical signal decomposition and reconstruction, a feature learning method based on the spherical Haar wavelet is proposed. In order to obtain the salient features of the spherical signal, a rotating spherical Haar wavelet is also proposed, which has a consistent scale in the same direction between the reconstructed image and the original image. Second, in order to find a better characteristic representation of the spherical signal, a higher contractive autoencoder (HCAE) is designed for the potential representation of the spherical Haar wavelet coefficients, which has two penalty items, respectively, from Jacobian and two order items from Taylor expansion of the point x for the contract learning of sample space. Third, in order to improve the classification performance, this paper proposes a fuzzy Gaussian support vector machine (FGSVM) as the top classification tool of the deep learning model, which can punish some Gaussian noise from the output of the deep HCAE network (DHCAEN). Finally, a DHCAEN-FGSVM classifier is proposed to identify the location of the pupil center. The experimental results of the public data set and actual data show that our model is an effective method for spherical signal detection.

As a group of complex neurodevelopmental disorders, autism spectrum disorder (ASD) has been reported to have a high overall prevalence, showing an unprecedented spurt since 2000. Due to the unclear pathomechanism of ASD, it is challenging to diagnose individuals with ASD merely based on clinical observations. Without additional support of biochemical markers, the difficulty of diagnosis could impact therapeutic decisions and, therefore, lead to delayed treatments. Recently, accumulating evidence have shown that both genetic abnormalities and chemical toxicants play important roles in the onset of ASD. In this work, a new multilabel classification (MLC) model is proposed to identify the autistic risk genes and toxic chemicals on a large-scale data set. We first construct the feature matrices and partially labeled networks for autistic risk genes and toxic chemicals from multiple heterogeneous biological databases. Based on both global and local measure metrics, the simulation experiments demonstrate that the proposed model achieves superior classification performance in comparison with the other state-of-the-art MLC methods. Through manual validation with existing studies, 60% and 50% out of the top-20 predicted risk genes are confirmed to have associations with ASD and autistic disorder, respectively. To the best of our knowledge, this is the first computational tool to identify ASD-related risk genes and toxic chemicals, which could lead to better therapeutic decisions of ASD.

Workload assessment faces two major issues. That is, how to learn effective fatigue characteristics and how to find the potential state of the workload. This paper proposes a solution to assess the brain fatigue workload of pilots through an instantaneous spectral entropy feature and an infinitely warped model. The instantaneous characteristics of electroencephalography (EEG) signals are extracted by Hilbert transform, and Euclidean norm weighted permutation entropy is proposed. The infinitely warped model is a new automatic learning model for detecting arbitrary shapes of EEG data. In addition, we propose a rapid learning framework to learn mental fatigue by integrating Treelet transform and infinitely warped models. Compared to other state-of-the-art methods, our approach is better able to handle complex data in complex shapes. The experimental results show that this method can more effectively assess the brain fatigue of pilots.

Large skin parts play an important role in the aerospace industry. The wall thickness of the machined pocket in the skin part needs to be strictly controlled to ensure the transport capacity and structural strength. The wall-thickness accuracy is generally decreased by various factors, such as the shaping error of the workpiece blank, fixing error, machine tool error, and deformation caused by cutting force or internal stress. These factors are usually inevitable and stochastic due to the extremely weak rigidity and easy-to-deflect characteristics of the large skin parts. To ensure the wall-thickness accuracy, a fuzzy v-support vector machine (Fv-SVM)-based wall-thickness error decomposition method is proposed. The wall-thickness errors, which are monitored in the cutting process, are decomposed into spatial-related errors and time-related errors. The Fv-SVM-based decomposition method with the principle of spatial statistical analysis is a data-driven approach for intelligent manufacturing. The data-driven method can consider all factors that affect the wall-thickness accuracy, while the model-driven method usually only considers one factor, such as the workpiece deformation or fixing error. After decomposition, the spatial-related wall-thickness error is offline compensated, and the time-related wall-thickness error is compensated by using a real-time strategy. The novel method can be applied to complex tool paths. The cutting experiment of rectangular pockets in a large skin panel was conducted to verify the effectiveness of the proposed method. The wall-thickness accuracy can be improved to 0.05 mm for the workpiece with only 2 mm thickness.

In this paper, a novel bacterial foraging algorithm (BFA)-Gaussian support vector classifier machine (GSVCM) model was proposed to improve the fatigue classification accuracy of electromyography (EMG) signals. This optimization mechanism involves the kernel parameter setting in the GSVCM training procedure, which significantly influences the classification accuracy. Experiments were conducted based on the EMG signal to differentiate the normal and fatigue status. In the proposed method, the EMG signals were decomposed into intrinsic mode functions by ensemble empirical mode decomposition (EEMD) before the mean instantaneous frequency could be obtained by Hilbert transform (HT). Finally, the fatigue statistical features can be extracted from fast Fourier transform and EEMD-HT. The application of this model to the fatigue status recognition of EMG signal indicated that further significant enhancement of the classification accuracy can be achieved by the proposed BFA-GSVCM classification system. The diagnostic method is effective and feasible.