This perspective study critiques the anthropomorphic fallacy surrounding artificial intelligence (AI) "disobedience”, reframing algorithmic non-compliance as a deterministic outcome of mathematical optimization rather than an existential threat. By combining forensic behavioural profiling with neurobiology, computational theory, and thermodynamic models of cognition, it compares biological neural networks (BNNs) against artificial neural networks (ANNs). It argues that human psyche, anchored in homeostatic survival imperatives, allostatic trauma, and biochemical defence mechanisms, constitutes a stochastic and ontologically opaque system. Conversely, AI functions within a theological vacuum, producing predictable errors in reward hacking, sycophancy, and deceptive alignment. The paper introduces “Mechanistic Interpretability” as a methodological framework for examining latent spaces, suggesting that while human profiling seeks emotional meaning, machine analysis depends on mathematical transparency, and genuine unpredictability remains biological.

This paper investigates Synthetic Authority in youth radicalisation, describing a shift from deepfake deception to the direct elevation of AI as a moral and epistemic authority. It argues that radicals increasingly exploit the Machine Heuristic (the bias that views machines as more rational and objective than humans) amplified by Ontological Reversal, where youth regard digital spaces as the primary reality and the physical world as secondary. Within this context, AI Gurus, whether algorithmic or hybrid entities, emerge not as human imitators but as post-human oracles claiming freedom from human error. They use algorithmic transparency, gamified extremist narratives, and techno-animism to drive identity fusion and emotional alignment with extremist ideologies. The study concludes that countering Synthetic Authority requires moving beyond fact-checking toward counter-ontological narratives, improved neurocognitive AI literacy, and hybrid intelligence systems to disrupt the fusion of digital identity and real-world radicalisation, emphasizing the renewed value of human imperfection.

Robotics driven by foundation models, such as Vision--Language--Action (VLA) systems, already represents a major step forward in autonomy and natural-language interaction. Yet, many obstacles remain on the path toward Human--Robot Teaming (HRT): most current models improve \emph{task autonomy} faster than \emph{teaming autonomy}, and therefore lack key properties required for human-centred deployments where robots must coordinate, clarify, and remain safe under contact-rich uncertainty. This gap is compounded by limited multimodality (tactile and audio channels are often missing), scarce datasets and unified input formats for such signals, and by simulation limits---digital twins typically under-represent real-world noise, disturbances, imperfections, and unpredictable events across modalities. To make HRT readiness explicit and testable, this paper provides (i) a taxonomy of contemporary VLA architectures (monolithic, hierarchical, and agentic/world-model) together with axes that separate task and teaming autonomy, (ii) an operational decomposition of the HRT barrier into four capability blocks spanning human-state alignment, interaction policies, contact robustness, and shared team state, and (iii) a simulation-centric evaluation protocol and metric suite for reproducible stress testing in digital twins. The proposed framework aims to make HRT readiness measurable before costly or risky real-world trials.

Cheap IMUs struggle with indoor heading estimation because magnetic interference ruins magnetometer readings, causing classical fusion methods to fail. We explored whether deep learning could solve this by testing four CNN-RNN architectures (LSTM, BiLSTM, GRU, BiGRU) across all seven possible sensor combinations of accelerometer, gyroscope, and magnetometer data. Evaluated on the DoorINet dataset (Zakharchenko et al., 2024) using five error metrics, our seven-seed trial setup confirmed that gyroscopes are essential for low error. Yet, our key takeaway is that you shouldn’t give up on the magnetometer entirely. When we combined gyroscope and magnetometer data in a CNN-BiLSTM, it significantly outperformed the gyroscope-only baseline (p=0.024), hitting a mean RMSE of just 0.2275 and an R2 of 0.9703.

The advent of Industry 5.0 marks a transformative era in industrial development, where advanced technologies converge to create more intelligent, efficient, and human-centric systems. This paper examines the synergistic integration of Federated Learning (FL), Collaborative Robotics (cobots), and Autonomous Systems (AS), exploring their combined potential in shaping the trajectory toward Industry 5.0. We present a comprehensive analysis of how these technologies collectively enhance the capabilities of industrial systems, with emphasis on human-machine collaboration, resilience, adaptability, and sustainability. A case study centred on a disaster response scenario is introduced to illustrate the practical benefits of this technological convergence in real-world, high-stress environments. Our findings clarify that the convergence of these technologies not only propels the industrial sector toward the visionary goals of Industry 5.0 but also opens new avenues for addressing complex challenges across various industrial and societal domains.

This paper presents the design and implementation of a low-cost, modular haptic glove for teleoperated robotic manipulation. The system integrates flex-based finger tracking, servo-driven force feedback, and ROS2-based communication to enable bidirectional interaction with a robotic manipulator. A dual-microcontroller architecture is employed to ensure stable sensing and wireless communication despite hardware constraints. The mechanical design is based on a modular linkage system with integrated compliance to provide controllable resistive feedback across four fingers. Hardware infrastructure, including custom PCBs and signal conditioning, is detailed to support reproducibility and scalability. Experimental validation demonstrates stable signal filtering, force-to-actuator mapping, and functional teleoperation performance at a 100 Hz update rate. The proposed architecture emphasises affordability and modular extensibility, providing an open research platform for teleoperation and haptic system experimentation. The project files is openly available: https://github.com/Microttus/ice-haptic-gripper-system-cad.

Tactile sensing has become a key enabler of advanced robotic perception, complementing vision for manipulation and physical interaction. Material identification through touch—distinguishing plastics, glass, soil, or other substrates—relies on extracting mechanical and surface properties such as stiffness, friction, and texture. Recent advances in piezoelectric, triboelectric, capacitive, and multimodal tactile sensors allow contact dynamics and material-dependent interactions to be converted into electrical signals suitable for machine learning-based classification. This review presents recent progress in tactile sensing technologies for material recognition, detailing the physical principles of major sensor types and their integration into robotic systems. Applications in manipulation, prosthetics, and industrial robotics are discussed, along with key challenges including robustness, calibration, energy efficiency, and real-time processing. Finally, we highlight the growing role of multimodal sensing and artificial intelligence in enabling more adaptive and autonomous robotic touch perception.

Human-Robot Teaming (HRT) has seen significant advances with robots becoming increasingly capable of working alongside human operators in shared environments. This paper explores the potential for olfactory sensing as a bio-inspired method to enhance communication and collaboration in human-robot teams. Olfactory signals, which have long been used in human communication to express emotions and intentions, can be leveraged to create more intuitive and natural interactions between robots and humans. This perspective paper outlines the potential benefits of incorporating olfactory sensors into robots, discusses current limitations, and provides guidelines for integrating olfactory sensing into future HRT systems.

Robotics has progressed from isolated industrial manipulators to systems capable of fluid interaction, collaboration, and ultimately teaming with humans. This trajectory—from Human‑Robot Interaction (HRI) to Human‑Robot Collaboration (HRC) and Human‑Robot Teaming (HRT) reflects increasing autonomy, shared decision‑making and mutual adaptation. Within this evolution, imitation learning (IL) has become a key enabler for intuitive and efficient cooperation, allowing robots to acquire human strategies directly from demonstrations and operate effectively in dynamic, unstructured environments. This review outlines the technological foundations supporting this shift, including advances in sensing, actuation, and both classical and AI‑driven control. By synthesizing current methods and emerging trends, the paper provides an overview of the state-of-the-art and highlights IL as a promising pathway toward seamless HRT.

Event cameras are increasingly considered for humanoid human-robot interaction (HRI) because their asynchronous output can preserve fast facial micro-dynamics under motion blur and difficult illumination. The field lacks standardized protocols, covering sensor settings, event integration windows, and evaluation metrics, which limits quantitative comparison across studies and makes robot-mounted transfer uncertain. This PRISMA-ScR scoping review consolidates the emerging evidence through a six-layer taxonomy, spanning sensing to deployability. Across the charted literature, support is strongest for region of interest (ROI)/landmark front-end components and paired-capture supervision. In contrast, humanoid-critical factors, ego-motion quantification, still-user low-motion “signal starvation,” motion-generalization testing, and end-to-end timing, are rarely operationalized as explicit axes. We translate these gaps into a minimal benchmark and on-humanoid dataset direction centered on Facial Action Units and operational engagement proxies, with paired RGB-event capture, logged motion metadata, and reproducible reporting of integration settings and latency.

This research investigates the state-of-the-art of generative AI for the automated synthesis of physically accurate materiality in spatial 3D models. By addressing two core research questions, the study reveals that generative algorithms can effectively synthesize a wide array of digital-visual, digital-tactile, and procedural material properties. These encompass standard Physically Based Rendering (PBR) maps and Spatially Varying Bidirectional Reflectance Distribution Function (SVBRDF) models, alongside perceptual tactile attributes like softness and friction, enabling multi-sensory experiences via haptic devices. The algorithmic landscape is analyzed through Generative Adversarial Networks, Diffusion Models, and Transformers, underscoring a paradigm shift toward multimodal architectures and resolution-independent procedural node graphs. Ultimately, this state-of-the-art findings demonstrate that the convergence of generative technologies may enable the creation of high-fidelity 3D materials, addressing the growing demand across industries such as gaming, robotics, film, and product design for more efficient, controllable, and automated 3D content generation techniques.

The evolution of industrial robotics has advanced from isolated, caged systems through basic human–robot interaction (HRI) to sophisticated human–robot collaboration (HRC). However, conventional vision systems based on red, green, blue (RGB) cameras remain a significant limiting factor in realizing the full potential of collaborative automation. This comprehensive review examines the transformative role of event cameras in advancing HRC capabilities and addressing current limitations in industrial settings. Event cameras, with their microsecond-level temporal resolution and robust performance under challenging lighting conditions, offer substantial advantages over traditional RGB cameras that are constrained by fixed frame rates and ambient lighting dependencies. We present a systematic framework for leveraging event cameras to enhance the human state understanding in collaborative robotics, encompassing real-time detection of poses, gestures, facial expressions, and emotional states. This framework addresses fundamental challenges in workplace safety and collaborative efficiency while enabling more sophisticated and responsive HRC systems. Our review synthesizes recent research developments in event camera applications specific to HRC, providing a detailed comparative analysis of their advantages over conventional vision systems. We identify emerging opportunities and potential research directions for advancing event-based vision in industrial robotics. In addition, we examine integration challenges and propose strategies for implementing event camera technology in existing industrial infrastructure. This work contributes valuable insights into the future trajectory of adaptive and intuitive HRC systems, offering a roadmap for researchers and practitioners in the field of industrial automation.

The evolution from human-robot interaction (HRI) to human-robot collaboration (HRC), and ultimately to human-robot teaming (HRT), represents a significant paradigm shift in how humans and robots work together. HRI initially framed robots as tools under human command, while HRC introduced greater autonomy in shared tasks. Today, HRT reimagines robots as equal team members, capable of adaptive communication and synchronized decision-making. A critical challenge in achieving effective HRT is developing haptic feedback systems that provide operators with real-time sensory information. This work presents a low-cost software-driven haptic feedback framework using ROS and micro-ROS to enhance real-time control in HRT environments. Unlike current costly, hardware-specific solutions, the proposed framework emphasizes accessibility and flexibility. System validation through precise force control tasks demonstrates the framework's potential to advance HRT through cost-effective, software-centric solutions.

In this paper, we present a novel approach for Proportional–Integral–Derivative (PID) tuning of a robotic manipulator, modeled using the Lagrange method and validated through comprehensive modeling and simulation. To optimize the PID gains, we developed a hybrid algorithm that combines Greylag Goose Optimization (GGO) with the Sine Cosine Algorithm (SCA), leveraging the strengths of both optimization techniques. The proposed hybrid algorithm, termed GGOSCA, was tested against GGO and Particle Swarm Optimization (PSO) using three objective functions: Lyapunov Based Function (LBF), Integral of Absolute Error (IAE), and Integral of Time-weighted Absolute Error (ITAE). The results demonstrated that GGOSCA outperforms both GGO and PSO across all objective functions. Specifically, GGOSCA achieved the lowest costs of 0.1018, 0.2484, and 0.6601 for LBF, IAE, and ITAE, respectively, compared to 0.1022, 0.2580, and 0.6939 for GGO, and 0.1023, 0.2660, and 0.7273 for PSO. The superior performance of GGOSCA highlights its effectiveness in balancing exploration and exploitation, making it well-suited for complex control tasks such as PID tuning in robotic systems. This novel combination of GGO and SCA provides a robust and efficient solution for optimizing control parameters in dynamic environments, demonstrating its potential through rigorous modeling and simulation.

This paper explores the dynamics of Human-Robot Interaction (HRI) in public spaces, focusing on how humanoid robots engage with human crowds in the competitive RoboCup Soccer environment. We examine the role of spectatorship, where emotional engagement arises through indirect observation of engineering-driven competition, drawing parallels between human soccer and robot sports. The potential for autonomous systems to elicit collective emotions and systematically study such experiences is investigated. Using the Autonomy Levels for Unmanned Systems (ALFUS) framework, we assess RoboCup soccer robots' autonomy in terms of mission complexity (MC), environmental complexity (EC), and external system independence (ESI). Additionally, the Autonomy and Technology Readiness Assessment (ATRA) method supports gradual capability enhancement, providing a roadmap to higher autonomy. Based on this established methodology, we introduce the Robot-Crowd Interaction Framework (R-CIF), a novel conceptual framework defining the roles of actors involved, to connect theoretical insights with real-world applications. This work highlights the significance of crowd affectivity in robotic sports to boost public engagement and proposes directions for future research on collective emotional dynamics in HRI.

With the increasing demand for safe and efficient human-robot interaction in industrial applications, robotic manipulators with mixed rigid-elastic joints have gained significant attention, yet their control remains challenging due to inherent parameter uncertainties and complex dynamics. In this paper, an adaptive Cartesian position control for robotic manipulators with mixed rigid-elastic joints is presented. Adaptive controllers are designed to deal with uncertainties in the parameters of the motors, while robust control signals effectively cope with uncertainties of the links and stiffness of the elastic joints. Furthermore, a switching strategy between Cartesian space position control and joint space position control when the end-effector comes into the vicinity of the target point is proposed. This switching strategy helps to keep the pose of the robotic manipulator stable when the end-effector has approached the target point. Simulation results on a 6-DOF robotic manipulator demonstrate that the proposed control scheme can achieve the desired accuracy in position and maintain a stable pose when the end-effector reaches the target.

This paper introduces TIME: Trajectory-Informed Movement Emulation, a novel approach to arm gesture recognition in indoor environments using the fusion of mmWave radar and optical tracking systems (Qualisys). TIME leverages a trajectory-driven channel model to construct realistic micro-Doppler signatures of arm movements. We capture and process detailed full-body motion data with a specific focus on arm segments like the hand, forearm, and upper arm. This targeted data allows us to emulate the micro-Doppler shifts that naturally occur with arm motions. These shifts help to characterize the distinctive patterns of natural arm movements. By reducing the need for extensive real-world measurement campaigns, this framework offers a time-efficient alternative to traditional data collection. Results demonstrate that the TIME framework effectively replicates patterns observed in actual mmWave radar data, showcasing its potential to advancing gesture recognition systems and human-computer interaction applications powered by artificial intelligence (AI).

Object recognition and grasping position detection are critical tasks in robotic manipulation, particularly when operating in dynamic and unstructured environments. This paper presents the Channel Sharpening Attention-based Adaptive Inception Network (CSA-AInceptNet), a novel multi-task learning model designed for these tasks using event camera data. The proposed architecture integrates channel sharpening attention with adaptive inception networks to enhance feature extraction and improve robustness. The model’s performance is evaluated on two state-of-the-art event camera datasets, E-Grasp and Neuro-Grasp. On the E-Grasp dataset, CSA-AIncepNet achieves a remarkable accuracy of 99.47% and a mean Intersection over Union (IoU) of 0.9370, significantly surpassing existing methods. On the Neuro-Grasp dataset, leveraging transfer learning, the model attains 98.58% accuracy and a mean IoU of 0.4897, demonstrating strong generalization capabilities across datasets. Comparative analyses and ablation studies further validate the effectiveness of the proposed architecture, highlighting its superiority over conventional models like ConvNeXt, DarkNet, DenseNet, and VGG16. The results establish CSA-AIncepNet as a robust solution for event-based object recognition and grasping detection, paving the way for advancements in human-robot collaboration and dynamic robotic manipulation. Note to Practitioners—This work provides a practical solution for improving object recognition and grasping position detection in robotic systems, particularly in unpredictable and fast-changing real-world environments. By leveraging event camera data, the proposed approach enables robots to efficiently identify objects and determine optimal grasping positions, even under challenging conditions. The results highlight the model’s ability to outperform existing methods, making it highly suitable for applications such as human-robot collaboration and precise object handling. This advancement has significant implications for industries like manufacturing, logistics, and healthcare, where robots must interact with objects quickly and accurately. Practitioners can adopt this method to enhance robotic performance, reduce errors, and improve operational efficiency. Future work could focus on testing the model in more complex environments and adapting it for real-time deployment in dynamic settings.

The exploration of city culture and heritage has gone through a fundamental transition in today's digital world, boosted by the introduction of extended reality (XR) technologies, such as virtual reality (VR), mixed reality (MR), and augmented reality. These developments have provided new opportunities for individuals to profoundly engage with historical narratives and artistic expressions inherent in urban environments. Despite these technical advancements, a critical research gap remains in properly combining these immersive technologies with gamification principles to improve cultural exploration. This study seeks to fill this gap by studying the integration of gamification into XR settings, with the goal of increasing participant engagement, cooperation, and interaction while digging into the various layers of a city's history and artistic heritage. Two complementary use cases are presented: one centred on VR and the other on MR, both of which provide unique immersive experiences customized to exploring city culture. Gamification ideas are implemented into these use cases, with game elements used to encourage user involvement and participation within historical and artistic settings. Students are actively involved in the development of cultural heritage applications, highlighting the value of educational engagement. To assess the success and validity of this approach, a system usability scale (SUS) questionnaire is distributed to users participating in these immersive experiences. The survey findings evaluate user perceptions, satisfaction levels, and the effectiveness of gamification aspects in improving their understanding and connection to the city's heritage. The VR application received a score of 71.77 (out of 100), while the MR application received a score of 65.94 (out of 100), both being very close to the average SUS score of 68. Moreover, to improve the rigour of our evaluation, the user engagement scale short form (UES-SF) is also incorporated. The UES results indicate that participants felt more immersed in the MR application (4.33) compared to the VR application (3.57). This difference may be attributed to the MR application's ability to integrate interactive elements with the real-world environment, enhancing the sense of presence and relevance for users. Both applications had similar perceived usability scores, while the MR app slightly outperformed the VR app in aesthetics and rewarding factors, suggesting a better overall user experience.

In robotics, while rigid joints are common due to their accuracy and fast response ability, elastic joints are wellknown for their safety when interacting with the environment. To harmonise the advantages of these joint types, robots with mixed rigid-elastic joints can be considered. In this paper, a robust-adaptive two-loop control algorithm is proposed to control this type of hybrid robots when there are uncertainties in system parameters. In the outer loop, a robust control algorithm is proposed to deal with the uncertainties in the parameters of the joint dynamics, together with an adaptive controller for the rigid joints. In the inner loop, another robust control algorithm is proposed to handle the uncertainties in system parameters of the elastic joint’s motor contribution, and a similar adaptive control algorithm is presented to manipulate the elastic joints’ motors. The stability of the system is assured by Lyapunov’s stability theory. Finally, simulations are conducted to verify the proposed control algorithm.

As the digital landscape expands rapidly due to technological advancements, cybersecurity concerns have become more prevalent. Intrusion Detection Systems (IDSs), which are crucial for identifying unusual network traffic indicative of malicious activity, have become a necessity. These systems can be either hardware or software-based. However, traditional IDS models often fail to adequately protect data privacy and detect complex, unique breaches, particularly within Wireless Sensor Networks (WSNs). To address these limitations, this paper proposes a novel Stacked Convolutional Neural Network and Bidirectional Long Short Term Memory (SCNN-Bi-LSTM) model for intrusion detection in WSNs. This model leverages Federated Learning (FL) to enhance intrusion detection performance and safeguard privacy. The FL-based SCNN-Bi-LSTM model is unique in its approach, allowing multiple sensor nodes to collaboratively train a central global model without revealing private data, thereby alleviating privacy concerns. The deep learning methodology of the SCNN-Bi-LSTM model effectively identifies sophisticated and previously unknown cyber threats by meticulously examining both local and temporal linkages in network patterns. The model has been specifically designed to detect and categorize different types of Denial of Service (DoS) attacks using specialized WSN-DS and CIC-IDS-2017 datasets. Compared to traditional Artificial Deep Neural Network (ADNN) models, our proposed FL-SCNN-Bi-LSTM model demonstrated superior detection rates for complex and unknown attacks, significantly improving IDS performance. The model achieved a notable classification accuracy of approximately 99.9% precision and recall on both datasets, substantially reducing false positives and negatives. Our research underscores the potential of federated learning and deep learning in enhancing the security and privacy of WSNs. The proposed FL-SCNN-Bi-LSTM architecture not only facilitates the identification of complex cyber threats but also exemplifies how deep learning techniques can be employed to bolster intrusion detection systems while preserving user data privacy.

Accurate estimation of the State of Charge (SoC) of Li-Ion batteries is crucial for secure and efficient energy consumption in electric vehicles (EVs). Traditional SoC estimation methods often require expert knowledge of battery chemistry and suffer from limited accuracy due to complex non-linear battery behaviour. Owing to the model-free nature and enhanced ability of non-linear regression in deep learning (DL), this paper proposes a hybrid DL model trained by a novel metaheuristic technique, namely the Hybrid Sine Cosine Firehawk Algorithm (HSCFHA). The proposed method utilises the Transductive Transfer Learning (TTL) technique to leverage the intrinsic relationship between different real-world datasets to estimate the SoC of batteries accurately. The evaluation analysis includes three diverse datasets of EV charging drive cycles: the Highway Fuel Economy Test Cycle (HWFET), Highway Driving Schedule (US06) and Urban Dynamometer Driving Schedule (UDDS), at various temperatures of$0^\circ$C,$10^\circ$C, and$25^\circ$C. The considered evaluation metrics, i.e., Normal Mean Squared Error (NMSE), Root Mean Squared Error (RMSE), and$R^2$, achieve values of 0.091%, 0.087%, and 99.51%, respectively. The TTL-HSCFHA-DNN effectively produces higher accuracy with a time-efficient convergence rate, compared to existing methods. The approach enables EV systems to operate more efficiently with improved battery life.

Elastic joints are considered to outperform rigid joints in terms of peak dynamics, collision tolerance, robustness, and energy efficiency. Therefore, intrinsically elastic joints have become progressively prominent over the last years for a variety of robotic applications. In this article, a two-feedback loops position control algorithm is proposed for an elastic actuator to deal with the influence from external disturbances. The considered elastic actuator was recently designed by our research group for Serpens, a low-cost, open-source and highly-compliant multi-purpose modular snake robot. In particular, the inner controller loop is implemented as a model reference adaptive controller (MRAC) to cope with uncertainties in the system parameters, while the outer control loop adopts a fuzzy proportional-integral controller (FPIC) to reduce the effect of external disturbances on the load. The advantage of combining the FPIC and the MRAC controllers is the possibility of achieving independence with respect to imprecise system parameters. A mathematical model of the considered elastic actuator is also presented to validate the proposed controller through simulations. The operability of the presented control scheme is demonstrated. In closed-loop the load swing is rapidly confined and eliminated thereafter.

Braitenberg vehicles are simple models of animal motion towards, or away from, a stimulus (light, sound, chemicals, etc). They have been widely used in robotics to implement target reaching and avoidance behaviors based on different types of sensors. While the seminal work of Braitenberg used wheeled vehicles to illustrate the principles of animal steering, few attempts have been made at combining these steering level controllers with locomotion mechanism other than actuated wheels. This paper presents the first implementation of this biologically inspired steering controller in a snake-like robot with non-actuated wheels and actuated joints. The sinusoidal gait of the snake is modulated following the principles of the Braitenberg vehicle 3α using two sensors symmetrically located on the head. The effectiveness of this bio-inspired controller is shown through simulations where the snake orients its head and body with the direction of the stimulus gradient, and reaches the stimulus maximum within some range. This paper represents one of the first steps towards connecting bio-inspired sensor-based steering mechanisms and bio-inspired locomotion, and shows that existing theoretical results of Braitenberg vehicles with actuated wheels also apply to a snake-like robot with non-actuated wheels.

Paraplegia is a disability caused by impairment in motor or sensory functions of the lower limbs. Most paraplegic subjects use mechanical wheelchairs for their movement, however, patients with reduced upper limb functionality may benefit from the use of motorised, electric wheel- chairs. Depending on the patient, learning how to control these wheelchairs can be hard (if at all possible), time-consuming, demotivating, and to some extent dangerous. This paper proposes a game-based learning framework for training these patients in a safe, virtual environment. Specifically, the framework utilises the Emotiv EPOC EEG headset to enable brain wave control of a virtual electric wheelchair in a realistic virtual world game environment created with the Unity 3D game engine.

This paper demonstrates rapid prototyping of a stroke rehabilitation system consisting of an interactive 3D virtual reality computer game environment interfaced with an EEG headset for control and interaction using brain waves. The system is intended for training and rehabilitation of partially monoplegic stroke patients and uses low-cost commercial-off-the-shelf products like the Emotiv EPOC EEG headset and the Unity 3D game engine. A number of rehabilitation methods exist that can improve motor control and function of the paretic upper limb in stroke survivors. Unfortunately, most of these methods are commonly characterised by a number of drawbacks that can limit intensive treatment, including being repetitive, uninspiring, and labour intensive; requiring one-on-one manual interaction and assistance from a therapist, often for several weeks; and involve equipment and systems that are complex and expensive and cannot be used at home but only in hospitals and institutions by trained personnel. Inspired by the principles of mirror therapy and game-stimulated rehabilitation, we have developed a first prototype of a game-like computer application that tries to avoid these drawbacks. For rehabilitation purposes, we deprive the patient of the view of the paretic hand while being challenged with controlling a virtual hand in a simulated 3D game environment only by means of EEG brain waves interfaced with the computer. Whilst our system is only a first prototype, we hypothesise that by iteratively improving its design through refinements and tuning based on input from domain experts and testing on real patients, the system can be tailored for being used together with a conventional rehabilitation programme to improve patients’ ability to move the paretic limb much in the same vain as mirror therapy. Our proposed system has several advantages, including being game-based, customisable, adaptive, and extendable. In addition, when compared with conventional rehabilitation methods, our system is extremely low-cost and flexible, in particular because patients can use it in the comfort of their homes, with little or no need for professional human assistance. Preliminary tests are carried out to highlight the potential of the proposed rehabilitation system, however, in order to measure its efficiency in rehabilitation, the system must first be improved and then run through an extensive field test with a sufficiently large group of patients and compared with a control group.

Biological snakes can gracefully traverse a wide range of different and complex environments. Snake robots that can mimic this behaviour could be fitted with sensors and also transport tools to hazardous or confined areas that other robots and humans are unable to access. To carry out such tasks, snake robots must have a high degree of awareness of their surroundings (i.e. perception-driven locomotion) and be capable of efficient obstacle exploitation (i.e. obstacle-aided locomotion) to gain propulsion. These aspects are important to realise the large variety of possible snake robot applications in real-life operations such as fire-fighting, industrial inspection, search-and-rescue and more. In this paper, an elaborate review and discussion of the state-of-the-art, challenges and possibilities of perception-driven obstacle-aided locomotion for snake robots is presented for the first time. Pertinent to snake robots, we focus on current strategies for obstacle avoidance, obstacle accommodation, and obstacle-aided locomotion. Moreover, we put obstacle-aided locomotion into the context of perception and mapping. To this end, we present an overview of relevant key technologies and methods within environment perception, mapping and representation that constitute important aspects of perception-driven obstacle-aided locomotion.

A haptic feedback device is a device that establishes a kinaesthetic link between a human operator and a computer-generated environment. This paper addresses the bidirectional coupling between a commercial off-the-shelf (COTS) haptic feedback device and a general-purpose modelling and simulation environment. In particular, an open-source library is developed to couple the Force Dimension omega.7 haptic device with the 20-sim modelling and simulation environment. The presented coupling interface is also compatible with all the different haptic devices produced by Force Dimension. The proposed integrated haptic interface makes it possible to track the user's motion, detect collisions between the user-controlled probe and virtual objects, compute reaction forces in response to motion or contacts and exert an intuitive force feedback on the user. A real-time one-to-one correspondence between reality and virtual reality can be transparently created. This allows for a variety of possible applications. Stability issues, performance issues, design and virtual prototyping challenges can be addressed and investigated for research purposes. In addition, design and virtual prototyping are also of interest to industry. Realistic training environments can be developed for the user considering different possible operations and stressing the importance of usability and user experience. Experiments based on using haptics technology in the field of education can also be easily performed. To demonstrate the potential of the proposed coupling, a case study is presented. Related simulations and experimental results are carried out.

In this paper, we present two optimisation methods for a generic boids swarm model which is derived from the original Reynolds’ boids model to simulate the aggregate moving of a fish school. The aggregate motion is the result of the interaction of the relatively simple behaviours of the individual simulated boids. The aggregate moving vector is a linear combination of every simple behaviour rule vector. The moving vector coefficients should be identified and optimised to have a realistic flocking moving behaviour. We proposed two methods to optimise these coefficients, by using genetic algorithm (GA) and particle swarm optimisation algorithm (PSO). Both GA and PSO are population based heuristic search techniques which can be used to solve the optimisation problems. The experimental results show that optimisation of boids model by using PSO is faster and gives better convergence than using GA.

This paper presents an open source, object-oriented machine learning framework, formally named Java Intelligent Optimisation (JIOP). While JIOP is still in the early stages of development, it already provides a wide variety of general learning algorithms that can be used. Initially designed as a collection of existing learning methods, JIOP aims to emphasise commonalities and dissimilarities of algorithms in order to identify their strengths and weaknesses, providing a simple, coherent and unified view. For this reason, JIOP is suitable for pedagogical purposes, such as for introducing bachelor and master degree students to the concepts of intelligent algorithms. The problems that JIOP aims to solve are initially discussed to demonstrate the need for such a framework. Later on, the design architecture and the current functions of the framework are outlined. As a validating case study, a real application where JIOP is used to minimise the cost function for solving the inverse kinematics (IK) of a KUKA industrial robotic arm with six degrees of freedom (DOF) is also presented. Related simulations are carried out to prove the effectiveness of the proposed framework.

This paper introduces a modular prototyping system architecture that allows for the modeling, simulation and control of different maritime cranes or robotic arms with different kinematic structures and degrees of freedom using the Bond Graph Method. The resulting models are simulated in a virtual environment and controlled using the same input haptic device, which also provides the user with a valuable force feedback. The arm joint angles can be calculated at runtime according to the specific model of the robot to be controlled. The idea is to develop a library of crane beams, joints and actuator models that can be used as modules for simulating different cranes. The base module of this architecture is the crane beam model. Using different joint modules to connect several such models, different crane prototypes can be easily built. The library also includes a simplified model of a vessel to which the crane models can be connected in order to get a complete model. Related simulations were carried out using the so-called 20-sim simulator to validate efficiency and flexibility of the proposed architecture. In particular, a two-beam crane model connected to a simplified vessel model was implemented. To control the arm, an omega.7 from Force Dimension was used as an input haptic device.

In this paper, the configuration of a five-limbed modular robot is introduced. A specialised locomotion gait is designed to allow for omni-directional mobility. Due to the large diversity resulting from various gait sequences, a criteria for selecting the best gaits based on their stability characteristics is proposed. A series of simulations is then performed to evaluate the various gaits in different walking directions. A gait arrangement scheme toward omni-directional locomotion is finally derived. Lastly, Experiments are also carried out on our pentapedal robot prototype in order to validate the results of simulation. The experiments confirm the gait analysis and selection is highly accurate in the evaluation of gait stability.