Solving the inverse kinematics problem of redundant robots operating in complex environments via a modified genetic algorithm

摘要

A new approach to solve the inverse kinematics problem of redundant robot manipulators in environments cluttered with obstacles is presented in this paper. The physical problem is formulated as an optimization problem under constraints, and solved via a modified genetic algorithm (mGA). The mGA searches for successive robot configurations in the entire free space so that the robot moves its end-effector from an initial placement to a final desired. The objective of this optimization is tosimultaneously minimize the end-effector's positional error and the robot's joint displacements. The efficiency of the proposed approach is demonstrated through multiple experiments carried out on several redundant robot manipulators. Comparison with twoother approaches, the well known Pseudoinverse method and a previous method based on a simple GA, shows that the accuracy of the present solution is substantially better.