Adaptation to the dynamic environment and variable task sequence is the critical ability for robot navigation and task execution. The Cyclic Networking Rapidly-exploring Random Tree (CNRRT) method is proposed to obtain the optimal path in real time and realize long-term path planning ability in a complex dynamic environment. The cyclic branch is introduced to the acyclic graph of Rapidly-exploring Random Tree (RRT) method, which forms a decentralized path network in the configuration space. An iterative searching strategy is built to search for the optimal path in the network. The branch prune, reconnection, and regrowth processes enable the decentralized network to efficiently respond to dynamic changes in the environment. The CNRRT can search for the real-time optimal path in the dynamic environment, dealing with the configuration and task changes robustly. Besides, the CNRRT is consistent for scenarios with long-term task sequence without significant performance fluctuation. Simulations and real-world comparative experiments verify the effectiveness of the proposed method.

A major challenge in laryngeal surgery today is the limited flexibility of surgical operations. To address the limitation, this paper proposes a novel continuum robot (CR) system with enhanced dexterity, a robust inverse kinematics algorithm, and a sensorless automatic calibration method. The proposed CR possesses 4 flexible degrees of freedom, allowing for control based on angles and end-effector position. Compared with traditional Jacobian-based methods, the proposed inverse kinematics algorithm effectively addresses the singularity issue arising from curvature hypotheses. Mitigating the singularity is crucial for ensuring continuous and stable motion planning. The calibration method enables automatic initialization without additional sensors, a capability not previously reported in the literature. The efficient and automatic calibration reduces preparation time for laryngeal surgery. Compared to the manual calibration, which requires approximately 210 s, the proposed method reduces the calibration time by 160 s, thereby achieving a 76.19 % improvement in efficiency. Taking the damped least squares method as the baseline, the inverse kinematics algorithm reduces the maximum solving error from 7.36 mm to just 0.05 mm. Furthermore, the CR is capable of dexterous motion within narrow and curved cavities. The phantom and animal experiment results demonstrate the practicality and reliability of the proposed CR system in laryngeal surgery.

In recent years, there has been growing interest in developing robots and autonomous systems that can interact with human in a more natural and intuitive way. One of the key challenges in achieving this goal is to enable these systems to manipulate objects and tools in a manner that is similar to that of humans. In this paper, we propose a novel approach for learning human-style manipulation skills by using adversarial motion priors, which we name HMAMP. The approach leverages adversarial networks to model the complex dynamics of tool and object manipulation and the aim of the manipulation task. The discriminator is trained using a combination of real-world data and simulation data executed by the agent, which is designed to train a policy that generates realistic motion trajectories that match the statistical properties of human motion. We evaluated HMAMP on one challenging manipulation task: hammering, and the results indicate that HMAMP is capable of learning human-style manipulation skills that outperform current baseline methods. Additionally, we demonstrate that HMAMP has potential for real-world applications by performing real robot arm hammering tasks. In general, HMAMP represents a significant step towards developing robots and autonomous systems that can interact with humans in a more natural and intuitive way, by learning to manipulate tools and objects in a manner similar to how humans do.

In scenarios such as environmental data collection and traffic monitoring, timely responses to real-time situations are facilitated by persistently accessing nodes with revisiting constraints using unmanned aerial vehicles (UAVs). However, imbalanced task allocation may pose risks to the safety of UAVs and potentially lead to failures in monitoring tasks. For instance, continuous visits to nodes without replenishment may damage UAV batteries, while delays in recharging could result in missing task deadlines, ultimately causing task failures. Therefore, this study investigates the problem of achieving balanced multi-UAV path planning for persistent monitoring tasks, which has not been previously researched according to the authors’ knowledge. The main contribution of this study is the proposal of two novel indicators to assist in balancing task allocation regarding multi-UAV path planning for persistent monitoring. One of the indicators is namely the waiting factor, which reflects the urgency of a task node waiting to be accessed, and the other is the difficulty level which is introduced to measure the difficulty of tasks undertaken by a UAV. By minimizing differences in difficulty level among UAVs, we can ensure equilibrium in task allocation. For a single UAV, the ant colony initialized genetic algorithm (ACIGA) has been proposed to plan its path and obtain its difficulty level. For multiple UAVs, the K-means clustering algorithm has been improved based on difficulty levels to achieve balanced task allocation. Simulation experiments demonstrated that the difficulty level could effectively reflect the difficulty of tasks and that the proposed algorithms could enable UAVs to achieve balanced task allocation.

In recent years, the removal of orbital debris has become an increasingly urgent task due to advancements in human space exploration. Capturing space debris through caging manipulation offers notable advantages in terms of robustness and controllability. This paper proposes a configuration-based caging manipulation design method for a cable-driven flexible arm. First, the cable-driven flexible arm with multi self-lockable links is introduced. To quantify the caging configurations formed by different self-lockable link selections, a novel planar caging quality metric is then presented for arbitrary planar objects. Using this metric, a caging design method is developed for the flexible arm to conduct caging manipulations. Finally, the caging manipulation strategies are discussed with lock selection maps for different objects, and a robust caging strategy considering uncertainty is further explored. Simulation and experimental results validate the effectiveness of the proposed caging design method and demonstrate better performance compared to conventional caging methods.

The selection of random sampling points is crucial for the path quality generated by probabilistic roadmap (PRM) algorithm. Increasing the number of sampling points can enhance path quality. However, it may also lead to extended convergence time and reduced computational efficiency. Therefore, an improved probabilistic roadmap algorithm (TL-PRM) is proposed based on topological discrimination and lazy collision. TL-PRM algorithm first generates a circular grid area among start and goal points. Then, it constructs topological nodes. Subsequently, elliptical sampling areas are created between each pair of adjacent topological nodes. Random sampling points are generated within these areas. These sampling points are interconnected using a layer connection strategy. An initial path is generated using a delayed collision strategy. The path is then adjusted by modifying the nodes on the convex outer edges to avoid obstacles. Finally, a reconnection strategy is employed to optimize the path. This reduces the number of path waypoints. In dynamic environments, TL-PRM algorithm employs pose adjustment strategies for semi-static and dynamic obstacles. It can use either the same or opposite pose adjustments to avoid dynamic obstacles. Experimental results indicate that TL-PRM algorithm reduces the average number of generated sampling points by 70.9% and average computation time by 62.1% compared with PRM* and PRM-Astar algorithms. In winding and narrow passage maps, TL-PRM algorithm significantly decreases the number of sampling points and shortens convergence time. In dynamic environments, the algorithm can adjust its pose orientation in real time. This allows it to safely reach the goal point. TL-PRM algorithm provides an effective solution for reducing the generation of sampling points in PRM algorithm.

With increasing age, many elderly individuals will not be able to stand normally. To solve this problem, a knee exoskeleton is designed. The knee joint is designed as a variable stiffness structure. It can adjust its stiffness according to the body’s movement state, ensuring precise assistance while also enhancing human comfort. The variable stiffness mechanism consists of an elastic output actuator and a stiffness-adjusting actuator. The elastic output actuator is mainly responsible for the output of the joint torque. The stiffness-adjusting actuator is mainly responsible for adjusting the joint stiffness. These two mechanisms are analysed separately. Based on their relationship with the whole mechanism, a stiffness model of the entire knee joint is established. Experiments are subsequently conducted to evaluate the variable stiffness joint. The stiffness identification experiment indicates that the actual stiffness of the whole knee joint is essentially consistent with the theoretical value. The trajectory tracking experiment demonstrates that the joint exhibits excellent trajectory tracking capability, although stiffness has a certain effect. The exoskeleton assistive effect experiment demonstrates the ability of the exoskeleton to assist in standing. Additionally, the experiment on subjects with exoskeletons of different stiffnesses determines the impact of stiffness on human comfort.

Biped wall-climbing robots (BWCRs) serve as viable alternatives to human workers for inspection and maintenance tasks within three-dimensional (3D) curtain wall environments. However, autonomous climbing in such environments presents significant challenges, particularly related to localization and navigation. This paper presents a pioneering navigation framework tailored for BWCRs to navigate through 3D curtain wall environments. The framework comprises three essential stages: Building Information Model (BIM)-based map extraction, 3D climbing path planning (based on our previous work), and path tracking. An algorithm is developed to extract a detailed 3D map from the BIM, including structural elements such as walls, frames, and ArUco markers. This generated map is input into a proposed path planner to compute a viable climbing motion. For path tracking during actual climbing, an ArUco marker-based global localization method is introduced to estimate the pose of the robot, enabling adjustments to the target foothold by comparing desired and actual poses. The conducted experiments validate the feasibility and efficacy of the proposed navigation framework and associated algorithms, aiming to enhance the autonomous climbing capability of BWCRs.

This study introduces an innovative deep learning method for intelligent healthcare emotion analysis, specifically targeting the recognition of pain based on facial expressions. The suggested approach combines cloud-based mobile applications, utilising separate front-end and back-end elements to optimise data processing. The main contributions consist of a Smart Automated System (SASys) that integrates statistical and deep learning methods to extract features, thereby guaranteeing both resilience and efficiency. Image preprocessing encompasses the tasks of detecting faces and normalising them, which is crucial for extracting features with high accuracy. The comparison of statistical feature representation using Histogram of Oriented Gradients and Local Binary Pattern, along with machine learning classifiers, against an enhanced deep learning-based approach with an integrated multi-tasking feature known as multi-task convolutional neural network, demonstrates encouraging outcomes that support the superiority of the convolutional neural network architecture. Statistical and deep learning-based classification scores, when combined, greatly enhance the system’s overall performance. The results of the experiments prove that the method is effective, outperforming traditional classifiers and exhibiting comparable accuracy to cutting-edge healthcare SASys.

Although deep reinforcement learning (DRL) techniques have been extensively studied in the field of robotic manipulators, there is limited research on directly mapping the output of policy functions to the joint space of manipulators. This paper proposes a motion planning scheme for redundant manipulators to avoid obstacles based on DRL, considering the actual shapes of obstacles in the environment. This scheme not only accomplishes the path planning task for the end-effector but also enables autonomous obstacle avoidance while obtaining the joint trajectories of the manipulator. First, a reinforcement learning framework based on the joint space is proposed. This framework uses the joint accelerations of the manipulator to calculate the Cartesian coordinates of the end-effector through forward kinematics, thereby performing end-to-end path planning for the end-effector. Second, the distance between all the linkages of the manipulator and irregular obstacles is calculated in real time based on the Gilbert–Johnson–Keerthi distance algorithm. The reward function containing joint acceleration is constructed with this distance to realize the obstacle avoidance task of the redundant manipulator. Finally, simulations and physical experiments were conducted on a 7-degree-of-freedom manipulator, demonstrating that the proposed scheme can generate efficient and collision-free trajectories in environments with irregular obstacles, effectively avoiding collisions.

Spherical robots face significant challenges in motion control on non-horizontal terrains, such as slopes, due to their unique spherical structure. This paper systematically investigates the motion stability of spherical robots on inclined surfaces through modeling, control algorithm design, and experimental validation. Precise Equilibrium Modeling: Using the virtual displacement method, the precise equilibrium equation for spherical robots on slopes is derived, addressing the issue of insufficient accuracy in describing the actual center of gravity in existing studies. Control Algorithm Design: For known slope conditions, a Backstepping Control (BSC) algorithm is designed, demonstrating excellent tracking performance. For unknown slope conditions, an Adaptive Backstepping Control (ABSC) algorithm is proposed, which significantly reduces tracking errors and enhances system robustness through parameter adaptation. Simulation and Physical Validation: Simulations confirm the effectiveness of the algorithms: BSC achieves high-precision control under known slopes, while ABSC exhibits strong adaptability under unknown slopes. Physical experiments validate the stability of the algorithms in a $5^\circ$ slope environment, demonstrating reliable performance across different control angles.

In order to solve the problem of poor quality of paths generated by the traditional Q-RRT* algorithm and blind random search, an improved APF-QRRT* algorithm is proposed in this paper. The improved APF-QRRT* algorithm obtains a set of discrete critical path points connecting the start point and the end point by the Q-RRT* algorithm, and then fine-tunes the paths by using the local optimization capability of the APF to improve the smoothness and safety of the paths. The traditional Q-RRT* algorithm is improved, and the fast alternating expansion of two random trees is realized by introducing a bidirectional search strategy of two random trees and adopting node greedy expansion, where the nearest node of the tree is used as the reference for the expansion of this tree during the iterative process of path node generation. The experimental results show that the improved APF-QRRT* algorithm reduces the path planning time by 20.3%, the path length by 1.8%, the number of path nodes by 33.3%, and the number of sampling points by 23.6% compared with the standard APF-QRRT* algorithm in a complex environment. In this paper, a system test platform is constructed and utilized to carry out multi-AGV path planning experiments in real environments, and the experimental results show that the proposed hybrid path planning algorithm has good path planning effects.

In the context of an aging population and declining birth rates, the advantages of robotic-assisted training are becoming increasingly prominent. However, improving the adaptability and safety of assistive walking robots remains a critical challenge. Accurately identifying a user’s turning intent is essential for preventing dangerous situations such as falls or slips. As one of the core parameters of lower limb motion, foot rotation angles not only reflect the stability and coordination of gait but are also crucial for accurately predicting walking intentions, such as straight walking and turning. This study proposes a gated recurrent unit-based model for predicting foot rotation angles, driven by 3D visual data. By constructing a lower limb linkage model that includes foot joints and incorporating 3D foot rotation angle features, we develop a real-time algorithm for gait state prediction. This model enables accurate prediction of walking intentions, such as straight walking or turning, during walking and is experimentally validated using a robotic walker. The experimental results demonstrate the effectiveness of the proposed predictive model.

Indoor positioning systems (IPS) are essential for mobile robot navigation in environments where global positioning systems (GPS) are unavailable, such as hospitals, warehouses, and intelligent infrastructure. While current surveys may limit themselves to specific technologies or fail to provide practical application-specific details, this review summarizes IPS developments directed specifically towards mobile robotics. It examines and compares a breadth of approaches that vary across non-radio frequency, radio frequency, and hybrid sensor fusion systems, through the lens of performance metrics that include accuracy, delay, scalability, and cost. Distinctively, this work explores emerging innovations, including synthetic aperture radar (SAR), federated learning, and privacy-aware AI, which are reshaping the IPS landscape. The motivation stems from the’ increasing complexity and dynamic nature of indoor environments, where high-precision, real-time localization is essential for safety and efficiency. This literature review provides a new conceptual, cross-border pathway for research and implementation of IPS in mobile robotics, addressing both technical and application-related challenges in sectors related to healthcare, industry, and smart cities. The findings from the literature review allow early career researchers, industry knowledge workers, and stakeholders to provide secure societal, human, and economic integration of IPS with AI and IoT in safe expansions and scale-ups.

This study presents an innovative system for upper limb rehabilitation, combining a variable stiffness device, the ReHArm prototype, with a dynamic and engaging user interface, known as Arms Rehabilitation Management System. The proposed system offers a highly customisable approach to rehabilitation, ensuring real-time adaptability to patients’ specific needs while maintaining compactness and ease of use. Key features include a modular design allowing precise stiffness adjustments, a robust control architecture, and interactive rehabilitation phases designed to enhance user engagement. Extensive multidisciplinary analyses, including kinematic, dynamic, and structural evaluations, demonstrate the system’s ability to improve therapy effectiveness through tailored interaction and feedback. Validation tests demonstrated the prototype’s reliability and robustness, and initial usability assessments suggest its potential to improve rehabilitation outcomes. Further clinical studies involving patients will be necessary to fully evaluate its therapeutic effectiveness.

Extensible continuum robots (ECRs) offer distinct advantages over conventional continuum robots due to their ability to enhance workspace adaptability through length adjustments. This makes ECRs particularly promising for applications that require variable lengths involving the manipulation of objects in challenging environments, such as risky, cluttered, or confined. The development of ECRs necessitates careful consideration of mechanical structures, actuation methods, methods of stiffness variability, and control methods. The selection of papers is based on their relevance to ECRs within the period of 2010 to 2023 in the databases of Scopus and Web of Science. Distinguishing itself from other review papers, this paper aims to deliver a comprehensive and critical discussion about the advantages and disadvantages of ECRs concerning their mechanical structures, actuation methods, stiffness variability, and control methods. It is a beneficial resource for researchers and engineers interested in ECRs, providing essential insights to guide future developments in this field. Based on the literature, existing ECRs exhibit an inherent trade-off between flexibility and structural strength due to the absence of systematic design methods. Additionally, there is a lack of intelligent and effective controllers for achieving complex control performance and autonomous stiffness variability.

The inverse dynamics model of an industrial robot can predict and control the robot’s motion and torque output, improving its motion accuracy, efficiency, and adaptability. However, the existing inverse rigid body dynamics models still have some unmodelled residuals, and their calculation results differ significantly from the actual industrial robot conditions. The bootstrap aggregating (bagging) algorithm is combined with a long short-term memory network, the linear layer is introduced as the network optimization layer, and a compensation method of hybrid inverse dynamics model for robots based on the BLL residual prediction algorithm is proposed to meet the above needs. The BLL residual prediction algorithm framework is presented. Based on the rigid body inverse dynamics of the Newton–Euler method, the BLL residual prediction network is used to perform error compensation on the inverse dynamics model of the Franka robot. The experimental results show that the hybrid inverse dynamics model based on the BLL residual prediction algorithm can reduce the average residuals of the robot joint torque from 0.5651 N·m to 0.1096 N·m, which improves the accuracy of the inverse dynamics model compared with those of the rigid body inverse dynamics model. This study lays the foundation for performing more accurate operation tasks using industrial robots.

With advancements in industrial robot technology and the ongoing enhancements in control system performance, the demand for precise robot motion is increasing. Generally, an increased number of interpolation points enhances the precision of robot movement, but excessive points can lead to jittering and out-of-step issues. This paper investigates the relationship between the number of motion interpolation points and the response times of the control system and the robot’s terminal velocity, based on the theoretical calculation and experimental analysis of the limit interpolation points for the control system of a self-developed 6-DOF (Six Degree of Freedom) robot. The method for calculating limit interpolation points is refined using the least squares method, and equations are derived for different control system response time and robot’s terminal velocity reaction times. The validity of the prediction curves is verified through experimental analysis.