Current geometric-based path-planning methods for vertical parking in automated parking systems face issues such as stringent initial position requirements, insufficient clearance from garage boundaries, and discontinuous curvature. To address these problems, this paper proposes a parking path-planning method that integrates the Reeds–Shepp curve with an improved particle swarm optimization (PSO) algorithm. The parking process is decomposed into forward and backward segments, each generated using the Reeds–Shepp curve to enhance flexibility. The shift point, defined as the transition between the two segments, serves as the decision variable in the PSO framework. The fitness function jointly considers path length and deviation between the target pose and garage entry. A quadratic Bézier curve and a nonlinear adaptation mechanism are incorporated into the PSO to determine the optimal shift point, mitigating the risk of proximity to garage boundaries. The initial path is then generated using the Reeds–Shepp curve based on the initial pose, shift point, and target pose. After discretization, the path points are optimized via a nonlinear optimization formulation to eliminate curvature discontinuities at segment junctions. Experimental results demonstrate that the proposed method generates feasible, continuous-curvature parking paths across a variety of initial poses, outperforming conventional approaches in terms of flexibility and clearance.

The appeal of quadruped robots lies not only in their ability to mimic the diverse and agile locomotion of animals but also in their potential to integrate with human artistic expression to achieve complex multi-skill movements. Continuous multi-skill motion in quadruped robots requires the realization of diverse, continuous, and long-horizon behaviors, involving a broader state space and more complex motion generation and transitions. This presents significant challenges, including sparse rewards, incomplete data, long-horizon motion planning, and the design of fine-grained motion transitions. In this work, we categorize quadruped robot skills into three types: rhythmic motions, expressive motions, and high-dynamic motions and generate reference trajectories for each category using central pattern generators, animation design, and motion capture, respectively. We then design an asymmetric neural network architecture and employ an imitation–reinforcement learning algorithm to train policies for generating these three types of motions. By composing multiple motion skill trajectories, we avoid long-horizon motion planning; by leveraging reinforcement learning, we enable smooth and continuous skill transitions; and by introducing a two-stage reference state initialization curriculum, the robot is able to switch from arbitrary states to the target motion skill. Moreover, during training, the policy imitates only key characteristics of the reference motions rather than strictly tracking fixed trajectories, making it more robust. Finally, we achieve robust motion skill generation and seamless transitions on a quadruped robot equipped with a 6-DoF manipulator, validating the effectiveness and feasibility of the proposed multi-skill generation and transition method.

The increasing adoption of robotic manipulators in the industry has brought the issue of robotic device safety to the forefront, particularly in the context of collaborative robotics. Among the main challenges, the evaluation of impacts with operators and objects in the workspace represents a critical issue, as accurately quantifying the magnitude of collisions is a complex task. Although current regulations propose various models for impact assessment, the suggested approaches are often oversimplified and not easily applicable in complex scenarios. To overcome this limitation, this study proposes the development of an effective mass impact model that simulates the collision between two bodies as an interaction between two lumped masses. The model is developed for both two-dimensional and three-dimensional analyses. Furthermore, it is implemented in a generic form and using a lumped-parameter robot model. This formulation enables the creation of a highly versatile model capable of overcoming the lack of knowledge of inertial parameters of robots. The approach is experimentally validated by means of a SCARA robot that collides with a cart, demonstrating the model’s applicability and accuracy.

For large-scale problems, efficiently optimizing task allocation with local communication and computing resources is challenging. Existing algorithms often fall short in task allocation computation time. Inspired by Bayesian estimation, we propose a parametric heuristic algorithm for redundant agents that can effectively perform dynamic task allocation in distributed architectures. Firstly, the method includes a confidence proxy-driven posterior correction algorithm (CPPCA), which can iteratively satisfy the constraint of the number of agents for each task based on local communication. We also prove that CPPCA can strictly converge after the number of neighbors is larger than a certain value. Then, we design a greedy swapping algorithm (GSA) to reduce the total cost while satisfying the constraints. It is also proved that the algorithm can converge in a finite number of steps under the condition that the communication graph is connected. Finally, the feasibility of the method is verified by simulation experiments. It shows that the confidence proxy-driven posterior correction and greedy swapping (CPPC-GS) method has significant advantages over the baseline algorithms in solving large-scale problems under the premise of redundant agents. It also has good results in the face of dynamic task allocation. Furthermore, the virtual-real fusion experiment bridges the “modeling gap” by validating the algorithm’s resilience to real-world engineering constraints, such as non-ideal communication and hardware heterogeneity, ensuring a seamless transition from theoretical frameworks to robust practical deployment.

Upper limb rehabilitation exoskeletons face fundamental challenges achieving coordinated multi-joint assistance within portable configurations. Current cable-driven systems demonstrate effective single-joint support but lack coordinated shoulder-elbow capabilities due to anchor-point sensing constraints that limit workspace during simultaneous movements. To address these limitations, this paper presents a novel coordinated 3-Degree of Freedom (DOF) shoulder-elbow exosuit (3.3 kg) employing motor-proximal sensing architecture. Strategic load cell repositioning from anchor points to actuation unit locations eliminates spatial constraints, while geometric compensation algorithms maintain measurement accuracy, enabling coordinated assistance (shoulder flexion/extension, abduction/adduction, elbow flexion/extension) with preserved 0–$90^\circ$ kinematic workspace. Systematic development integrating product design specifications, multi-criteria decision-making, and biomechanical component dimensioning provided traceable design synthesis. Preliminary proof-of-concept tests with healthy participants (n = 5) provide an initial assessment of the system performance, demonstrating adequate sensing accuracy within the primary Activities of Daily Living (ADL) workspace, measurable muscle activation reduction, and preserved natural kinematics with minimal range-of-motion (ROM) constraint. An extensive experimental validation is out of the scope of this work. Results establish technical feasibility for coordinated cable-driven assistance, identifying requirements for future clinical translation.

This paper presents a comprehensive navigation strategy aimed at enabling autonomous and collision-free motion of automated guided vehicles (AGVs) in indoor environments. The work addresses limitations of conventional AGV systems which are often restricted by static routing and inadequate adaptability to dynamic obstacles. A framework on global–local motion planning is developed, optimising (i) a global path planner based on the Probabilistic Roadmap (PRM) algorithm, (ii) a local trajectory tracking using the Pure Pursuit (PP) controller, and (iii) a real-time obstacle avoidance method leveraging the vector field histogram (VFH) algorithm. The decision-making layer oversees the coordination between global routing and local manoeuvring, dynamically switching between planning and avoidance modes in response to environmental changes. The proposed framework is validated through real-world experiments on a physical AGV platform operating in a structured environment. Experimental results show that the proposed framework generates feasible and smooth paths, while PP-based trajectory tracking achieves high final positioning performance when supported by fused odometry. In free-path navigation, fused odometry reduced the average final percentage error to 0.61% and increased the average final positioning accuracy to 99.40%. In obstacle-path navigation, the integrated PP–VFH framework maintained high final positioning accuracy at 99.10% under varying obstacle configurations. The coordination of PRM, PP, and VFH enables smooth, adaptive, and safe navigation of the AGV within a structured workspace. The proposed framework contributes a unified, modular architecture that strategies to bridges global path planning, local motion control, and real-time reactive avoidance within a single system.

To address the problems of insufficient adaptability of photovoltaic (PV) cleaning robots in mountainous environments, where it is challenging to balance cleaning repeatability and cleaning integrity, this paper proposes a three-legged compound mobile mechanism and conducts optimization of its key parameters. First, the topological structure of the mechanism is designed based on the locomotion mode of inchworms. Combined with the requirements of PV panel cleaning tasks and the characteristics of cleaning methods, a compound motion mode integrating linear cleaning and rotational cleaning is adopted to determine the final structural configuration of the mechanism. Second, considering the relationship between the key parameters of the mechanism and the robot’s cleaning repeatability as well as cleaning integrity, a comprehensive performance index is constructed. The genetic algorithm is introduced for parameter optimization. Simulation results show that the optimal solution occurs at the larger step length and the maximum unilateral width of the cleaning mechanism within the valid interval. Finally, a prototype was fabricated for experiments. The results demonstrate that the robot can maintain stable adsorption on inclined and undulating PV panels, and the actual cleaning path is consistent with the theoretical trajectory, verifying the effectiveness of the design and the accuracy of the theoretical analysis in this paper.

The integration of large language models (LLMs) into KUKA industrial robots offers a promising pathway toward intuitive natural language control and a higher level of intelligence. However, existing methods face two primary challenges: their reliance on predefined motion primitives and the syntactic fragility imposed by the strict constraints of proprietary KUKA Robot Language (KRL). To address these challenges, we propose MIKI (multi-agent integrated KUKA interface), a decoupled dual-agent framework that bridges high-level reasoning and low-level syntactic verification for industrial robot programming. The controller agent incorporates a chain-of-thought (CoT) reasoning mechanism to systematically decompose abstract human instructions into logical intermediate steps. By leveraging a visual toolchain, it translates these reasoned plans into a sequence of executable actions, each embedded with precise world coordinates. Subsequently, the code generation agent employs a syntax-guided iterative refinement process, underpinned by an ANTLR4 parser, to transform these actions into verified and executable KRL code. Experimental results demonstrate that this closed-loop approach substantially improves the syntactic success rate of code generation from 0.61 to 0.89. Furthermore, the framework achieves an average success rate of 0.6 across five desktop manipulation tasks of varying complexity using a physical KUKA robotic arm, validating its effectiveness in bridging high-level cognitive reasoning with deterministic industrial execution.

This study introduces an improved gesture-based human–swarm interaction control system for drone swarms that unifies motion and formation control while addressing the critical challenge of neglecting the impact of faulty robots on task performance. Unlike conventional approaches that treat these functionalities separately, our integrated framework enables robust task execution in challenging environments. Using 26 hand gestures (18 formations and 8 flight commands) processed by a variety of deep learning models, including convolutional neural networks (e.g., ResNet101 and MobileNetV2) and recurrent neural network-based approaches such as long short-term memory (LSTM), bidirectional LSTM, gated recurrent unit (GRU), and bidirectional GRU (BiGRU), the system provides comprehensive yet accessible control, notably through the BiGRU network with 99.4% accuracy in real-time gesture recognition tasks. Deployment testing conducted using the CrazyFlyt simulation platform demonstrated the system’s robustness in maintaining user-intended formations with low positional error and stable convergence times, even in the presence of sensor noise and faulty robots. Statistical analysis revealed highly significant main and interaction effects ($p\lt 0.05$) for swarm scale and scenario on performance metrics, indicating that the impact of swarm size is dynamically contingent upon environmental conditions. Post hoc evaluations predominantly highlighted significant performance degradation in faulty-robot scenarios compared to normal or noisy conditions, with swarm scale critically modulating these effects.

To guarantee the safety and stability of cable-driven continuum robots (CDCRs) in multi-obstacle scenarios, this paper proposes a novel trajectory planning and tracking framework that integrates an artificial potential field (APF) based model predictive control (MPC) with a two-layer B-spline trajectory optimization framework. APF is incorporated as a penalty term in the objective function to enhance trajectory safety. Additionally, an improved whale optimization algorithm (IWOA) is introduced to circumvent the local optimal trap of MPC. Particularly, to improve tracking accuracy and stability, a two-layer B-spline optimization framework is proposed to enhance the smoothness of the planned trajectory. Simulation and experimental results show that the designed trajectory planning and tracking algorithm can generate collision-free trajectories and achieve high tracking accuracy in multi-obstacle environments.

Gravity compensation, widely employed in mechanisms such as manipulators and exoskeletons, serves to minimize unnecessary energy consumption. Enhancing the performance of gravity compensation mechanisms (GCMs) is therefore of significant practical importance. One promising approach is the development of novel design methodologies. Existing GCMs utilizing cable-pulley units exhibit limitations in terms of structural compactness and efficiency. To address these issues, this paper proposes an innovative X-type gravity compensation unit featuring an adjustable cable sequencing arrangement. The application of this unit to a passive waist exoskeleton (PWE) is further introduced. Given the potential influence of pulley size and positioning on the actual spring elongation, a pulley-angle error model is established to analyze the effect of varying pulley positions. Simulations confirm the validity of the model and demonstrate that pulley size is a non-negligible factor in system performance. Finally, an experimental platform is constructed to evaluate the performance of the PWE. Experimental results indicate a notable improvement in exoskeleton performance attributable to the unique design of the X-type unit.

Force/motion transmissibility serves as an essential performance metric in the kinematic analysis of parallel mechanisms. The initial step in evaluating force/motion transmissibility involves identifying the basis elements of twist and wrench spaces. In parallel mechanisms consisting of PRS, RPS, SPR, and PU limbs, the identification of the basis elements of a given limb is relatively straightforward, as the limb constraints can be evaluated as pure forces and torques. However, in limbs where the axes of the joints are offset, determining these basis elements becomes quite complex. Typically, the constraints of these limbs involve the interconversion between general screws and line vectors or couples. This paper focuses on a 3-PRRU parallel mechanism. Procedures for identifying the basis elements are investigated. Screw theory is used to elucidate the relationships between the variations in the basis elements and the configurations of the mechanism. Moreover, the influence of an offset between two revolute joint axes on the force/motion transmissibility is systematically analyzed. The results indicate that, compared to the case without offsets, the introduction of offsets enhances the force transmissibility of the mechanism.

A methodology for impedance control based on mechanical vibration theory was presented to select stiffness and damping to meet the desired dynamic response using modal analysis. This methodology provided profound insights into how changes in stiffness and damping matrices influence a robot’s dynamic response under impedance control, by determining natural frequencies and damping ratios in the modal space. In this paper, our analytical and experimental results challenge the claim of a recent paper stating that the use of a null-space projection does not affect the outcomes, which appears to be a claim that is valid based only on their specific choice of stiffness and damping matrices and robot configuration. Through experiments using a 7-DOF Franka Emika Panda robot, we demonstrate the influence of both stiffness/damping matrices and the weighting matrix of null-space control on the responses of robot manipulators with degrees of redundancy performing primary and secondary tasks. We also introduce and show insights of the benefits of switching between the statically and the dynamically consistent projection matrices, contributing to the broader understanding of the dynamic response of redundant robotic manipulators.

To address the limitation of maximum jump height constrained by the maximum torque of hip joint motors in bipedal wheeled robots during trajectory planning using the dual-mass linear spring model (DMLSM), this study aims to reduce the maximum required hip joint torque for achieving equivalent jump heights. Concurrently, reducing the maximum rotational speed of hip joint demanded by equivalent jumps is essential to minimize motor wear and enhance system stability. Building upon the dual-mass model, we propose a trajectory planning method based on hip joint motor state optimization (TPM-HJMSO). This method integrates a nonlinear spring model, cubic polynomial interpolation, and a min–max optimization approach with nonlinear constraints. Simulation results demonstrate two critical advances: TPM-HJMSO achieves 163.16% higher jump heights than the DMLSM under identical maximum hip torque constraints, while reducing maximum hip motor speed by 37.17% when attaining equivalent jump heights. These outcomes validate the method’s superior performance. Precise trajectory tracking is demonstrated through feedforward-plus-PD control in both Webots simulations and physical prototype experiments, verifying TPM-HJMSO’s effectiveness.

The spiral structure of interventional robots is widely utilized for vascular dredging procedures. Variations in the spiral structural parameters of these robots exert a significant impact on the hemodynamic parameters within the vessel during the interventional process; therefore, optimizing the design of these parameters constitutes a crucial prerequisite for ensuring safe diagnosis and treatment. Based on the bidirectional fluid-structure interaction (FSI) between pulsatile blood and elastic blood vessels, theoretical calculations and experimental measurements of the fluid velocity inside the vessel during robotic intervention were conducted. The results demonstrate that the trends and magnitudes of the calculated values are essentially consistent with those of the measured values. Furthermore, the range and variance methods in orthogonal design were applied to numerically investigate the effects of three key spiral parameters – spiral pitch, spiral groove depth, and thread profile – on hemodynamic and vascular parameters, including blood velocity, blood pressure, vascular wall shear stress, and vessel deformation. The findings indicate that spiral groove depth is the primary and most significant factor influencing blood velocity, blood pressure, and vessel deformation. In contrast, spiral pitch emerged as the main and critical factor affecting vascular wall shear stress. Meanwhile, thread profile is found to be a secondary factor with relatively minor impacts on all the aforementioned parameters. To achieve the objective of safe interventional diagnosis and treatment, it is recommended that the robot’s spiral structure be designed with a smaller spiral groove depth, a moderate spiral pitch, and a thread profile featuring strong cutting capability.

This study systematically addresses the design and aerodynamic optimization of tail configurations for bio-inspired flapping-wing robot. Based on a flapping-wing robot prototype, a theoretical model linking tail-size parameters to the pitch static stability margin was established. Six types of tail configurations — arc-shaped tail, triangular-shaped tail, swallow-shaped tail, webbed tail, T-shaped tail, and V-shaped tail — comprising 17 parametrically varied specimens were designed and fabricated. Through a high-precision decoupled wind-tunnel test platform (free stream velocity: 10 m/s, flapping frequency: 2.5 Hz), the three-axis aerodynamic moments (pitch, yaw, roll) under various actuation states were quantitatively measured. The influence of key geometric parameters such as characteristic width, opening angle, and control-surface area on handling and stability was thoroughly investigated. Experimental results show that the T-shaped tail (#53) performs best in pitch control moment and lateral-directional stability margin, with a peak pitch moment of 0.185 N·m — approximately 42 % higher than the baseline configuration — demonstrating the most outstanding overall performance. The V-shaped tail (#62) exhibits excellent lateral-directional control capability under differential actuation, achieving a roll moment of up to 0.106 N·m, albeit with pronounced control coupling effects. This research provides reliable experimental evidence and a theoretical reference for the configuration selection and parameter optimization of tails in flapping-wing robot, offering significant engineering guidance for enhancing flight quality and control effectiveness. In addition, this paper establishes a theoretical framework for preliminary tail size design based on pitch static stability and proposes a parametric design method for multi-configuration tails for flapping-wing robot. The research results not only provide a theoretical basis for tail parameter selection but also offer experimental references for tail configuration optimization and subsequent control-oriented system design.

Precision agriculture has significantly advanced in recent years and is now widely used to improve the quality and sustainability of agricultural products by applying the right treatment, in the right place, at the right time. This paper presents the development and testing of AGRIMARO.Q (AGRIcultural MAte RObot), a novel omnidirectional, three-wheeled robot designed for agricultural applications in structured environments such as greenhouses. The platform features a reconfigurable track-widening mechanism that enhances its versatility by allowing it to navigate crop rows of varying widths. The paper provides a detailed discussion of the robot’s design, focusing on the track adjustment system and steering wheel drive subsystems, as well as the kinematic model and low-level control systems. Then, experimental tests are presented to evaluate both the robot’s energy consumption and its behavior in executing certain trajectories, such as row traversal and inter-row transitions, in different terrain types, replicating typical greenhouse operations.

Spherical robots can conduct surveillance in hostile, cluttered environments without being damaged. Their protective shell can safely house sensors and sensitive components. The pendulum-based spherical robot provides a platform (yoke) for a camera to be housed inside the protective hull. However, lateral oscillations, also known as wobble, become prominent when operated at low speeds, leading to shaky camera feedback. These oscillations are caused by the coupling between the forward and steering motions due to nonholonomic constraints. Designing a controller to limit wobbling in these robots is challenging due to their underactuated nature. In this paper, we propose a model-based (wobble-free) controller to navigate a pendulum-steered spherical robot to perform wobble-free turning maneuvers consisting of circular arcs and straight lines. The model is developed using Lagrange-D’Alembert equations and accounts for the coupled forward and steering motions. Furthermore, we also propose quantifiable metrics for wobble by deriving expressions for radius of curvature, precession rate, wobble amplitude, and wobble frequency during circular motions and designed an input–output feedback linearization-based controller to control the robot’s heading direction and wobble. Additionally, we extend the controller’s capabilities to handle point-to-point navigation, enabling the robot to execute more complex and goal-oriented trajectories. This research ensures the robot maintains wobble-free motion across diverse paths, enhancing its navigational autonomy and the quality of camera feedback. Overall, the proposed controller enables a teleoperator to command a specific forward velocity and pendulum angle as per the desired turning radius or destination while effectively limiting the robot’s lateral oscillations.

This paper proposes a kinematically redundant wall-climbing robot (CR), which employs a kinematically redundant parallel mechanism (PM) as its main body, combined with a serial mechanism, to achieve wall-climbing motion and avoid the dangers associated with manual operation. The improved Grübler-Kutzbach formula is utilized to analyze the degrees of freedom (DOF) of the PM. Subsequently, a kinematic analysis covering position, velocity, singularity, and workspace is conducted to establish the theoretical basis for the robot’s motion planning. To cope with complex working environments, trajectory planning is carried out for the CR, and the complex path planning problem in space is addressed through hierarchical processing, including global path planning, local path planning, and single-step motion planning. After theoretical analysis, kinematic, singularity, workspace analysis, and trajectory planning method of the proposed CR were verified through simulation experiments, confirming the correctness of the kinematic modeling and trajectory solution. Finally, a physical prototype was built to verify the climbing capability of the robot.

The shipborne 3UPS-3SS parallel mechanism is widely used in ship barge operation, but it has problems such as large volume and poor force transmission and acceleration performance. In order to solve the above problems, the kinematic/dynamic performance analysis and multi-objective optimization of scale parameters were carried out on the mechanism. The vector method is used to construct the position inverse solution model and constraint equations, and the kinematic inverse solution and velocity expression are derived, and the velocity Jacobian matrix is obtained. The fast polar coordinate search method was used to plot the reachable workspace by MATLAB software, and the volume expression of the workspace of the mechanism was obtained. The analytical equation of velocity and acceleration of each moving part was established by the differential method; the mechanism dynamics model was established by combining the virtual work principle, and the condition number was calculated and visualized based on the average mass matrix. Based on the distribution of the condition number of the velocity Jacobian matrix and average mass matrix, the force transferability and dynamic dexterity of the mechanism are evaluated. In the end, the workspace volume of the mechanism increased by 7.14%, the average minimum singular value increased by 196.43%, and the dynamic dexterity decreased by 55.69%, which provided a theoretical basis for the control and practical application of the mechanism.