In this article, we propose a navigation approach based on a flocking algorithm for multiple agents that discover the environment with their sensors. The proposed method is able to move the agents in a flexible formation that preserves a pattern that minimizes a measure of energy, but is also able to avoid collision with obstacles while maintaining communication and visibility between agents. Control laws are combined with a new version of the planner called the Nearness Diagram (ND), extended to multiple agents. Two central ideas in the approach are: (1) the global leader of the formation can change, selecting as global leader one that has favorable conditions to reach the goal, (2) each agent locally follows another agent, the agent that is followed minimizes a cost related to the shortest path to the global leader in a graph. The method is intensively validated with several computer simulations, implemented in differential-drive robots. The results showed that the approach maintains communication formation and flexible enough to allow agents to line up when narrow passages appear in the environment.

To address the issues of path roughness, low search efficiency, and the difficulty of simultaneously considering Ackermann vehicle kinematic constraints in practical path planning, this study developed a coordinated global–local path planning framework tailored for Ackermann-steered robots. In the global planning stage, a bidirectional A* search framework was enhanced with a dynamic heuristic function, improving the geometric quality of the global path while maintaining search efficiency. Path smoothing was applied to further align the path with Ackermann kinematic constraints. In the local planning stage, the Timed Elastic Band (TEB) algorithm was adapted for Ackermann steering by incorporating minimum turning radius and steering-related constraints, ensuring that the generated trajectories were both collision-free and executable. Python-based simulations on constructed maps were used to compare different heuristic strategies in terms of path length and search efficiency. Results showed that the proposed global planning strategy reduced path length by 34.62% compared with the traditional bidirectional A* algorithm and improved search efficiency by approximately 2.1%. Furthermore, Robot Operating System (ROS) simulations and real-robot tests were conducted to analyze curvature characteristics, velocity variations, and path-tracking performance under different global–local planning combinations. The experimental results demonstrated that the proposed planning framework offered strong engineering practicality and provided a feasible solution for real-world Ackermann-steered robots.

Butterflies are excellent fliers in nature, flapping mode, which can ensure the completion of complex flight movements, including rapid turns, hovering, and forward and backward flights. In this paper, the Chinese yellow swallowtail (Papilio xuthus) was chosen as the bionic research subject. The flapping motion and trajectory function of P. xuthus during takeoff were obtained by the high-speed camera. On this basis, a finite element model of butterfly forewings was established, the flapping trajectory function was imported, and the transient takeoff motion of forewing was simulated by Ansys Fluent 2021R2. The aerodynamic characteristics at different angles of attack during the takeoff were obtained. The flow structure characteristics of the forewing of P. xuthus were analyzed by SIMULIA XFlow 2020. It shows that the lift and drag coefficient of the forewings during takeoff both exhibit a trend of increasing first and then decreasing as the angle of attack increases. The maximum average lift-to-drag ratio is achieved, when the angle of attack is 20°. The obtained wingtip trajectory, optimal angle of attack, and vorticity parameters can provide new design ideas for solving problems such as insufficient takeoff capability and difficult mode switching of flapping-wing microair vehicles design.

This work describes the design and validation of a one-size-fits-all assistive exoskeleton for the upper limb that is self-adjustable to the wearer subject. The assistance function is performed even if the joints of the exoskeleton are not aligned with the joints of the subject; therefore, it does not require personalized adjustments during the wearing phase. The device is composed of a distal articulated system with a prismatic interface towards the body segment and a flexible proximal architecture that shifts the actuation towards the pelvis to limit the alteration of the subject’s center of gravity. In vivo experiments in the laboratory demonstrate the ability to alleviate muscular effort, and home-based experiments in performing daily activities show excellent perceived usability and acceptability in an elderly population. Furthermore, a proposed biomechanical model estimates the ability of the exoskeleton to contain the joint constraint reactions in the subject during the assistance phase.

An underactuated nonholonomic autonomous robot with a Dubins car kinematics navigates through a planar environment that hosts an unpredictably moving rigid formation of point-wise targets. Both the number of the targets and their geometric configuration are unknown. The robot measures the relative positions of the targets within a limited sensing range but has no access to their velocities or other kinematic parameters. The robot has to approach the formation, and to subsequently move so that first, all targets are surrounded by the robot’s path, and second, a predefined distance to the currently nearest target is maintained. However, a lower bound on the turning radius of the robot typically makes perfect achievement of the second objective impossible since different targets become the nearest to the robot in the course of its movement. So a scheme for on-the-fly synthesis of a feasible and somewhat optimal approximation of the ideal circumnavigation path is offered. Based on this development, the paper presents a computationally efficient control law that accomplishes the mission as much as possible. The effectiveness of the advocated approach is justified by mathematically rigorous proofs of nonlocal convergence and is confirmed by simulation tests and experiments with real robots.

The period-$1$ gait is a stable, efficient, and predictable walking pattern for bipedal robots, which is crucial for achieving precise control, energy optimization, and maintaining balance in complex environments. This study investigates the control of chaotic and multi-period gaits in a bipedal robot with impulsive actuation, focusing on the stabilization of such gaits into period-1 gaits and the expansion of their stable parameter regions. Three main contributions are presented: (1) stabilizing chaotic gaits to period-$1$ gaits using the Ott–Grebogi–Yorke (OGY) and bifurcation control methods; (2) converting multi-period gaits (e.g., period-$2$ and period-$4$) into period-$1$ gaits via bifurcation control; and (3) extending the stable parameter range of period-$1$ gaits by optimizing pulse thrust parameters. The bifurcation control method, though energy-intensive, proves more reliable than OGY method in preventing falls. Analytical and numerical results demonstrate that adjusting pulse thrust parameter $C$ significantly expands the stability domain, enhancing robustness against disturbances and parameter variations. The bifurcation control method proposed in this study is implemented via a constant thrust increment and validated through numerical simulations, which makes it easy to be applied in controller design. It provides both a theoretical foundation and a practical control scheme for achieving stable and efficient gait design in bipedal robots with impulsive actuation.

Effective impact mitigation is a critical aspect of ensuring the stable motion of legged robots, especially in challenging environments. This work introduces a leg motion control system integrating nonlinear active compliance and impedance control with a balance controller. Designed for high-speed dynamic locomotion, the proposed system achieves superior impact absorption and precise leg position tracking without relying on passive spring mechanisms. The proposed nonlinear control, featuring state-dependent adaptive laws for stiffness and damping, outperforms linear alternatives in vibration suppression while maintaining tracking accuracy. A nonlinear PD parameter tuning method and optimized footstep locations are utilized for the balance controller to ensure quick recovery from external disturbances. Experimental validation on the SCIT Dog platform demonstrates exceptional cushioning capability during high-speed maneuvers and effective full-degree-of-freedom impact mitigation, including lateral disturbances. Crucially, this work advances beyond prior research by demonstrating active springless leg compliance/impedance control with formal stability guarantees and represents a fundamental locomotion layer that is essential for enabling future autonomous navigation capabilities in quadruped robots operating in unstructured environments.

This paper addresses the challenge of robotic path planning in high-temperature operational environments. Aiming at the limitation of traditional methods that rely on geometric information while ignoring thermal radiation data, a potential field method path planning based on temperature-enhanced point cloud maps is proposed. By fusing the data of LiDAR and thermal imager, point cloud maps with temperature attributes are generated. Based on the traditional potential field method, the potential field function of thermal radiation repulsion was introduced. The temperature gradient of the heat source is transformed into the potential energy constraint of path planning to achieve path generation in high-temperature areas. The experimental results show that, compared with the traditional artificial potential field method, this study provides a new technical solution for the safe operation of robots in a thermal environment.

Humanoid dual-arm robots face significant challenges in planning safe and humanoid motions during collaborative tasks due to overlapping workspaces and kinematic redundancy. This paper proposes a real-time joint adjustment planning method that optimizes dual-arm motion by defining a “joint discomfort” metric, which quantifies the deviation of joint angles from their comfort positions, thereby enhancing the naturalness and human-likeness of the motion. Leveraging the parallel computing capability of recurrent neural networks (RNNs) and the redundancy resolution of dual arms, our method dynamically adjusts joint configurations to minimize discomfort while integrating trajectory tracking and obstacle avoidance constraints. Predefined tasks and constraints are formulated as a Quadratic Programming (QP) problem, efficiently solved using an RNN-based approach. Numerical simulations and physical experiments on the Ginger robot – a dual-arm system with 7-degree-of-freedom (DOF) manipulators – validate the efficacy of the proposed planning method in three key aspects: (1) optimizing joint space utilization to enhance workspace flexibility, (2) adaptively regulating joint motions, and (3) achieving sub-millimeter tracking accuracy (position error ¡0.15 mm) under dynamically constrained scenarios.

This article presents a comprehensive technical survey on the mechanical design of multi-domain multirotors (MDMs) – unmanned aerial systems capable of operating across diverse environments, specifically aerial-terrestrial, aerial-aquatic, and triphibious domains. Owing to their capacity for seamless transition between distinct media, MDMs have garnered significant attention for their potential in critical applications such as disaster response and environmental monitoring. However, existing literature has not sufficiently addressed the inherent design complexities and structural challenges of MDMs, particularly regarding the integration of multi-domain propulsion, actuation, and locomotion mechanisms. To address this gap, this survey systematically classifies MDM designs based on operational modes and actuation systems. It highlights key mechanical innovations, including reconfigurable architectures and multi-propulsion strategies. Finally, the survey identifies current research limitations and outlines pivotal directions for future investigation.

Parking in narrow spaces presents significant challenges, often resulting in deviations between vehicle’s final parking state (FPS) and normal FPS. These deviations can cause partial intrusion into the target parking spots, increasing the risk of collisions in the parking process and potentially making the parking spots unusable. To address these issues, this paper proposes an optimization method for both the parking path and FPS in narrow and non-ideal vertical parking scenarios. Initially, a partition-based calculation method for the minimum distance between the vehicle and obstacles (DBVO) was developed to quantitatively assess the impact of intrusion on parking safety. Following this, a predefined geometric set of clothoids is used to smooth curvature discontinuities in the parking path, and a four-phase parking path pattern is devised. Subsequently, a preferred method for parking manners is proposed by analyzing the effects of parking direction and maneuvers on spatial requirements. Finally, a two-step parking path optimization method is presented with the framework integrating both online and offline calculations, using the minimum DBVO field and a predefined path pattern. Comparative experiments demonstrate that this method could enhance the proportion of available intrusion scenarios, increase the success rate of effective path acquisition, and improve the overall quality of parking paths.

Solid-state LiDARs (SSLs), due to their limited horizontal field of view (FoV), suffer from degeneracy in indoor environments. In response, we propose placing reflectors in degraded regions and incorporating reflector feature residuals into the LiDAR Odometry and Mapping (LOAM) framework. This is combined with an Extended Kalman Filter (EKF) to fuse odometry predictions, reflector observations, and LOAM measurements, effectively mitigating localization drift in the submap. By employing CostMap-Multilateration-based loop closure detection and keyframe selection strategies focused on high-reflection regions of interest (ROIs), our approach achieves robust submap-to-submap registration through a coarse-to-fine process. This ensures globally consistent map construction. To validate the proposed method, qualitative and quantitative evaluations were conducted in real-world environments. The results show that in degenerate indoor environments, the performance of our approach is comparable to, and even exceeds, that of state-of-the-art SLAM methods.

To address the challenges posed by highly repetitive structures, unstructured terrain, and pronounced sensor disturbances in agricultural environments, this paper proposes a simultaneous localization and mapping framework that integrates explicit spatial feature modeling with time-varying noise estimation. First, a curved-voxel-based intensity–geometry joint probability model is constructed to transform conventional point features into spatial features capable of effectively capturing local structural patterns. These features are then aligned by minimizing the joint probability distance using the normal distributions transform, thereby enhancing feature discriminability and registration stability in highly repetitive scenes. Second, a maximum a posteriori-based recursive noise estimator is designed. By employing a dual sub-filter architecture, the proposed estimator enables joint modeling and online optimization of the noise parameters associated with LiDAR odometry and the inertial measurement unit, thereby improving the system’s adaptability to terrain-induced perturbations and sensor uncertainties. Experimental results demonstrate that the proposed method achieves a mean relative translation error of 1.42% and a mean relative rotation error of 1.53°/100 m in agricultural scenarios. In addition, the average error in row spacing estimation derived from the reconstructed map is constrained within 0.071 m. Compared with baseline methods, the proposed system exhibits significant advantages in both pose estimation accuracy and agronomic structural perception capability.

Non-spherical wrist manipulators are widely used in industrial settings, yet they lack effective inverse kinematics solutions. Current methods primarily rely on numerical iteration, which suffers from limitations such as sensitivity to initial values and convergence issues. Meanwhile, algebraic elimination methods require complex elimination techniques to obtain analytical solutions, restricting their practical engineering application. To address the limitations of these existing approaches, this paper innovatively proposes a novel analytical method based on conformal geometric algebra theory. The core of this method lies in utilizing the geometric characteristics of the manipulator to establish a polynomial equation for one of the joint variables, after which the remaining joint angles are solved according to the joint angle solution methodology. The correctness of the proposed method was verified using three sets of joint angles. Additionally, a comparative analysis with the Dixon elimination method was conducted using 100 sets of arbitrary initial joint angles. Example validation demonstrates that the proposed method exhibits significant advantages in computation time, solution completeness, and numerical accuracy, thereby providing a new theoretical tool and practical framework for the inverse kinematics analysis of robotic manipulators.

This endeavor encompasses establishing a dynamic model for a constrained joint module system in collaborative robots, predicated upon the Udwadia-Kalaba equation and constraint-following methodologies. Additionally, a leakage-type adaptive robust controller is developed to assuage uncertainties and disturbances. The structure of the dynamic model is methodically crafted, accommodating uncertain parameters to delineate the behavioral dynamics of the system. Furthermore, a meticulous derivation of a second-order representation of the constraint equations is undertaken to facilitate precise boundaries of the limitations imposed by the system. The proposed controller demonstrates adeptness in tailoring its strategies to the characteristics of both known and unknown attributes of the system, deftly navigating the intrinsic uncertainties of the system. Systematic simulations and empirical analyses have been performed to authenticate the efficacy of the advocated approach in regulating the joint module system. The research outcomes not only augment comprehension of robust control techniques for joint module systems but also furnish invaluable insights to propel future advancements within this specialized domain.