面向在轨捕获的空间机器人路径规划与控制综述

A review on planning and control of space robots for on-orbit capture

  • 摘要: 随着航天技术的快速发展,空间机器人逐渐被广泛应用于空间碎片清除、故障航天器维修、失控卫星捕获等在轨服务中. 在这类任务中,空间机器人的末端执行器需要跟踪预设轨迹到达捕获点,完成空间目标的捕获以及后续操作. 然而,复杂的太空环境和空间机器人固有的动力学特性使路径规划和控制面临诸多挑战. 本文首先回顾了世界主要航天国家在空间机器人领域的研究现状,介绍了空间机器人基于广义雅可比矩阵(Generalized Jacobian matrix, GJM)的运动学模型和基于拉格朗日法的动力学模型,以说明空间机器人的动力学特性. 随后,本文结合在轨捕获任务需求、复杂太空环境及动力学特性,归纳了空间机器人路径规划和控制领域的最新研究进展,展示了地面试验技术及其在路径规划与控制验证中的研究成果. 最后,本文总结了当前空间机器人技术中存在的问题,并展望了未来可能的研究方向.

     

    Abstract: With the rapid development of space technology, space robots are playing an important role in on-orbit services, such as refueling, debris removal, and malfunctioning satellite repair. These space robots typically consist of a base spacecraft and n degrees_of_freedom (n-DOF) manipulators. The manipulators could be equipped with different end-effectors to capture space objects and perform various operations. In the precapture phase, space robots are required to follow a preplanned trajectory to the designated capture points. However, the dynamic characteristics of space robots introduce challenges in path planning. First, dynamic singularity might occur when solving the inverse kinematics in task-space path planning. Unlike fixed-base manipulators, the motion of the base spacecraft could be influenced by the manipulator’s movements due to dynamic coupling. Thus, base disturbance minimization becomes a key consideration while planning the trajectory of the manipulator. In addition to dynamic factors, practical concerns such as obstacle avoidance as well as input and velocity constraints should be satisfied to ensure the success of the mission. Furthermore, for some complicated space missions involving dual-arm space robots, coordination between two arms should be considered. Once the desired trajectory is planned, a tracking control strategy is designed to drive the end-effector to the capture points. However, the high nonlinearity, multidimensionality, and strong coupling of the space robot system increase the difficulty of tracking control. During on-orbit capture, the relative motion between the space robot and the target limits the available operational time, requiring the end-effector to quickly reach the capture points along the desired trajectory. In addition, the controller design should guarantee both steady-state and transient performance, such as slight overshoot, small steady-state error and short transient time. Working in a complex space environment, space robots are inevitably subjected to unknown external disturbances and parametric uncertainties, which could negatively affect the control accuracy and require compensation. Consequently, the proposed control strategy should ensure overall system stability, fast convergence, and strong robustness against disturbances and uncertainties. Before deployment, comprehensive ground-based verifications are important to evaluate the performance of the system. The effectiveness of ground-based verifications in accurately reflecting real space motions depends on the simulations of the space environment, particularly the microgravity environment. To address these technological challenges, this paper provides an overview of recent advancements in path planning and tracking control, with a focus on the inherent dynamic characteristics of space robots and the complexities of the space environment. The application of intelligent methods in multiobjective optimization and disturbance rejection control is also introduced in light of the rapid development of artificial intelligence. Additionally, relevant ground-based verification technologies are discussed. Finally, this paper presents the existing limitations and potential future developments in the field of space robots.

     

/

返回文章
返回