关键词: 工业机器人, 逆运动学求解, 目标函数, 改进量子粒子群优化算法

Abstract: Aiming at the problems of singular pose, non-unique solution and low solution precision in the inverse kinematics solution processes of general robots, an improved quantum particle swarm optimization algorithm was proposed. Firstly, the robot kinematics model was established by using the D-H parameter method, and the minimum pose errors at the end of the manipulators were the main optimization goal, and the constraints of the minimum joint angle changes before and after the movement and the smooth and continuous stroke were added, and the objective function was designed. Secondly, an IQPSO algorithm was designed by using the Levy flight strategy to improve the particle update method, nonlinear dynamic adjust the shrinkage and expansion factors, and using the variable weight method to calculate the optimal average position. Then, the simulation and comparison experiments of three algorithms（IQPSO,APSO,QPSO） were carried out by simulating two different solutions of single point pose and continuous trajectory. The results show that the IQPSO algorithm has the advantages of fast convergence speed and high solution accuracy; finally, the IQPSO algorithm was used in the body of the robot arm for physical verification. The results show that the trajectory composed of interpolation points obtained by the IQPSO algorithm is continuous and smooth, which further proves the stability and feasibility of the algorithm in practical motion control.

Key words: industrial robot, inverse kinematics solution, objective function, improved quantum particle swarm optimization(IQPSO) algorithm