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.

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.

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.

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.

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.

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.

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.