Genetic Algorithm for Solving the Inverse Kinematics Problem for General 6R Robots

摘要

The problem of inverse kinematics for general 6R robots was provided for calculation process, and also exist many inverse kinematics and geometric structure which did not meet the PIEPER criterion. In order to solve these problems, an inverse kinematics algorithm with high accuracy based on multiple population genetic algorithm (MPGA) was proposed. Multiple population was performed to accelerate the convergence rate and avoid the defect of the least part point. For illustrating the performance of the MPGA, the simulation results attained from MPGA are compared with those obtained from well-known single-population genetic algorithm (SGA). Experiments on Panasonic TA1400 robot verified that the algorithm could calculate all globally optimal solutions of general geometric structure and the pose error also can have up to two digits after the decimal point. So this algorithm can be used to guarantee higher control accuracy.