Multi-objective configuration optimization for coordinated capture of dual-arm space robot

Abstract

Dual-arm space robot is a promising autonomous system to remove the increasing space debris actively. Due to its limited on-orbit energy, dual-arm space robot should approach and capture the target with minimum time and base disturbance. In this paper, the multi-objective configuration optimization is proposed for maximizing manipulability and minimizing base disturbance of dual-arm space robot during the pre-contact phase. Based on virtual base modelling and coordinated planning, the free-floating dual-arm space robots can be controlled to track and approach the target satellite. In addition, in the null-space of tracking motion, the configuration optimization of dual-arm space robot can be realized by arm-angle adjustment of 7 DOF redundant manipulator. Therefore, the expressions of arm angle and arm-angle Jacobian are first derived. Then the arm-angle trajectory is parameterized by adopting 5th order polynomial interpolation. The optimal arm-angle trajectory can be obtained by Multiple Objective Particle Swarm Optimization (MOPSO), in which the objective function of configuration optimization is set to the vector consisted of base disturbance and manipulability. Therefore, in the null space of dual-arm space robot approaching the tumbling target satellite, the configuration optimization is realized through following the optimal arm-angle trajectory. Finally, several single-objective and multi-objective optimization simulations are carried out to verify the proposed method.