Robot inverse acceleration solution based on hybrid genetic algorithm

Abstract

As the complexity of Jacobian matrix and the second order influence coefficient matrix of a robot, and the difficulty of calculating inverse Jacobian matrix, the problem of solving inverse acceleration is more difficult for a robot with degrees-of-freedom (DOFs) less than 6. Aimed at this problem, an approach for solving the robot inverse acceleration problem has been proposed in this paper, In which the hybrid genetic algorithm (HGA) and robot linkpsilas velocity and acceleration recursive formulas are employed to avoid calculation of inverse Jacobian matrix and the second order influence coefficient matrix. It is proved to be viable by practical computation of a 5-DOF robot inverse acceleration.