General Robot Inverse Kinematics Solution Based on MPGA-RBFNN

2018 3rd International Conference on Insulating Materials, Material Application and Electrical Engineering

材料科学;无线电电子学;电工学

Zhang, Yi^1 ; Liu, Fangjun^1

Chongqing University of Posts and Telecommunications, Chongqing

400065, China^1

关键词: Calculation accuracy;    Connection weights;    Kinematics modeling;    Multiple population genetic algorithms;    Network structures;    Nonlinear mappings;    Radial basis functions neural networks;    Robot inverse kinematics;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/452/4/042133/pdf
DOI  :  10.1088/1757-899X/452/4/042133

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

In order to solve the problem of the inverse kinematics of general robot, such as slow speed in problem-solving and lower accuracy of the solution, a high precision MPGA-RBFNN algorithm is proposed, which introduce the Multiple Population Genetic Algorithm (MPGA) into Radial Basis Functions Neural Network (RBFNN). Combining with the positive kinematics model of general robots, the RBFNN with three-layers is used to solve the inverse kinematics of general robots and the MPGA is adopted to optimize the network structure and connection weights of RBFNN. By using the way of hybrid coding and simultaneous evolutionary, the non-linear mapping from the posture of the robot in the working space to the angle of the joint is realized, avoiding the complicated formula derivation and improving the speed of solving. Finally, the experimental are tested on the 6R robot, the results show that the MPGA-RBFNN algorithm is not only improves the speed of the solving, but also enhances the training success rate and the calculation accuracy.