The dynamic factor particle swarm optimization (DFPSO) to inverse kinematics of CMOR based on pose decomposition strategy

Abstract

The Chinese Fusion Engineering Testing Reactor (CFETR) is a superconducting Tokamak device, whose components would get gradually damaged by being susceptible to high temperature, strong magnetic field, radiation, and so on, and should be maintained regularly. The CFETR Multipurpose Overload Robot (CMOR) is a redundant robot designed as a macro-micro arm structure for a gigantic workspace in the blanket and diverter of CFETR's maintenance. Due to the macro-micro arm structure, the robot has a high degree of freedom, and its serial DOF is up to 16 so the traditional algorithm is low efficient for the robot's kinematics. Hence, in this paper, a kinematic decomposition strategy is proposed to allocate or decouple the position aim. Besides, a novel particle swarm optimization (PSO) with dynamic accuracy called dynamic factor particle swarm optimization (DFPSO) is designed to accelerate solving the high-dimensional nonlinear kinematic problem and to improve accuracy. The results of the research show the validation and efficiency of DFPSO in the CMOR system's kinematics. Moreover, it's also a universal kinematic algorithm for all high-DOF robots.