In view of the fact that to get the analytic solutions of the inverse kinematics of robots must meet special conditions, this paper used a general solution, simulated annealing particle swarm optimization algorithm, to obtain the optimization of inverse kinematics without any special conditions. | Journal of Automation and Control Engineering Vol. 4, No. 4, August 2016 Inverse Kinematics of Multi-joint Robot Based on Particle Swarm Optimization Algorithm Ze Fan Cai 1,2 , Dao Ping Huang 2, and Yi Qi Liu 2 1 Electronics and Information Engineering Department of Shunde Polytechnic, Foshan, China 2 Automation College of South China University of Technology, Guangzhou, China Email: 12266423@, audhuang@, 653830373@ Abstract—The optimization of the inverse kinematics of robot is the most important part of robot kinematics. Parameters for each link of the robot are needed to be identified according to the position and orientation in both of teaching programming and interpolation. In view of the fact that to get the analytic solutions of the inverse kinematics of robots must meet special conditions, this paper used a general solution, simulated annealing particle swarm optimization algorithm, to obtain the optimization of inverse kinematics without any special conditions. Simulation results show that the algorithm can be effectively applied to the robot inverse kinematics and is able to achieve sufficient accuracy. Kennedy R. and C. Eberhart J in 1995. Because PSO is simple and easy to be realized, it has caused the attention of scholars, and hundreds of papers about PSO have been published. It has been applied successfully in many fields. [4] In this paper, simulated annealing mechanism is introduced to improve the standard PSO. This paper tries to obtain the optimization of inverse kinematics with this improved PSO. Simulation result shows that the algorithm can be effectively applied to the robot inverse kinematics and is able to achieve sufficient accuracy. II. Index Terms—particle swarm optimization algorithm, simulated annealing algorithm, multi-joint robot, inverse kinematics, optimization I. There are several robot pose expressions, such as Euler angle, rotation matrix, four element method, etc. The most common one is the .
