【摘要】 面向未来载人登月人类-机器人现场协同作业任务，开展了人机共融下机器臂作业过程安全轨迹规划研究，提出了基于机械臂动态调整机制和凸优化理论的轨迹规划方法。该方法充分考虑采样作业机械臂关节角速度、角加速度和转矩等多种约束，以时间和能耗为综合指标函数，将轨迹规划问题表征为含约束的数值优化问题；利用参数化方法将多维动力学约束简化成基于标量参数的约束，降低规划问题的计算复杂度；建立基于航天员和机器人相对运动学参数的安全评估机制，将航天员的安全融入规划问题，并利用直接配点法和McCormick包络方法求解并构建机械臂的安全轨迹。以机器人-人协同作业为背景进行算法仿真，结果表明本文所提出的安全轨迹规划能够实现在航天员干预下进行高效安全的机器人作业。更多还原

【Abstract】 Aiming at the future human-robot field cooperative operation task for the future manned lunar exploration missions, the safe trajectory planning of robot arm operation process under man-machine integration is studied, and a path planning method based on the dynamic adjustment mechanism of robotic arm and convex optimization theory is proposed. In this method, multiple constraints such as angular velocity, angular acceleration and torque are fully considered, and the path planning problem is characterized as an optimization problem with constraints by taking time and energy consumption as comprehensive index functions. In order to reduce the computational complexity of programming problems, the multidimensional dynamic constraints are simplified into scalar parameters based constraints by using parameterization method. A safety evaluation mechanism based on relative kinematic parameters of astronaut and robot is established. The safety of astronaut is integrated into the planning problem, and the safety path of manipulator is constructed by direct collocation method and quadratic cone optimization method. Simulation results show that the proposed safe path planning can realize efficient and safe robot sampling under astronaut intervention.更多还原