Space is the driving force of the world’s sustainable development, and ensuring the sustainability of human activity in space is also necessary. Robotic arm active debris capture removal (RA-ADCR) is a noteworthy technology for containing the dramatic increase in space debris and maintaining orbital safety. This review divides the RA-ADCR technology progress history into three periods and presents the status of related research. Two major development trends are summarized and subdivided through the analysis and collation of research achievements over the past three years. Taking the treatment of parameter uncertainties as the entry point, researchers would like to improve the discrimination accuracy and scope to reduce uncertainties. On the other hand, researchers accept such uncertainties and would like to offset and avoid the impact of uncertainties by extending the error margins. Subsequently, the challenges of RA-ADCR are analyzed in line with the task execution flow, which mainly focuses on the conflict between on-satellite computing power and the performance of task execution. In addition, feasible solutions for the current phase are discussed. Finally, future outlooks are evaluated and discussed.
6 May 2025 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
The anthropomorphic grasping capability of prosthetic hands is critical for enhancing user experience and functional efficiency. Existing prosthetic hands relying on brain-computer interfaces (BCI) and electromyography (EMG) face limitations in achieving natural grasping due to insufficient gesture adaptability and intent recognition. While vision systems enhance object perception, they lack dynamic human-like gesture control during grasping. To address these challenges, we propose a vision-powered prosthetic hand system that integrates two innovations. Spatial Geometry-based Gesture Mapping (SG-GM) dynamically models finger joint angles as polynomial functions of hand-object distance, derived from geometric features of human grasping sequences. These functions enable continuous anthropomorphic gesture transitions, mimicking natural hand movements. Motion Trajectory Regressionbased Grasping Intent Estimation (MTR-GIE) predicts user intent in multi-object environments by regressing wrist trajectories and spatially segmenting candidate objects. Experiments with eight daily objects demonstrated high anthropomorphism (similarity coefficient R2=0.911, root mean squared error RMSE=2.47°), rapid execution (3.0710.41 s), and robust success rates (95.43% single-object; 88.75% multi-object). The MTR-GIE achieved 94.35% intent estimation accuracy under varying object spacing. This work pioneers vision-driven dynamic gesture synthesis for prosthetics, eliminating dependency on invasive sensors and advancing real-world usability.
In recent years, due to the wide application of 3D vision in the fields of autonomous driving, robot navigation, and the protection of cultural heritage, 3D point cloud registration has received much attention. However, most current methods are time-consuming and are very sensitive to noises and outliers, resulting in low registration accuracy. Therefore, we propose a two-stage framework based on graph neural network and attention—TSGANet, which is effective in registering low-overlapping point cloud pairs and is robust to variable noises as well as outliers. Our method decomposes rigid transformation estimation into two stages: global estimation and fine-tuning. At the global estimation stage, multilayer perceptrons are employed to estimate a seven-dimensional vector representing rigid transformation directly from the fusion of two initial point cloud features. For the fine-tuning stage, we extract contextual information through an attentional graph neural network consisting of attention and feature-enhancing modules. A mismatch-suppression mechanism is also proposed and applied to keep our method robust to partially visible data with noises and outliers. Experiments show that our method yields a state-of-the-art performance on the ModelNet40 dataset.
The robot manipulator is commonly employed in the space station experiment cabinet for the disinfection task. The challenge lies in devising a motion trajectory for the robot manipulator that satisfies both performance criteria and constraints within the confined space of an experimental cabinet. To address this issue, this paper proposes a trajectory planning method in joint space. This method constructs the optimal trajectory by transforming the original problem into a constrained multi-objective optimization problem. This is then solved and integrated with the seventh-degree B-spline curve. The optimization algorithm utilizes an indicator-based adaptive differential evolution algorithm, enhanced with improved Tent chaotic mapping and opposition-based learning for population initialization. The method employed the Fréchet distance to design a trajectory selection strategy based on the Pareto solutions to ensure that the planned trajectory complies with Cartesian space requirements. This allows the robot manipulator end-effector to approximate the desired path in Cartesian space closely. The findings indicate that the proposed method can effectively design the robot manipulator trajectory, considering both joint motion performance and end-effector motion constraints. This ensures that the robot manipulator operates efficiently and safely within the experimental cabinet.