An Optimization Method for Inverse Kinematics of a 7-DOF Redundant Manipulator

Based on the D-H model of a normal 7-DOF redundant manipulator, a novel two-phase calculation optimization approach is proposed in this paper by combining the fixed joint-angle method with the weighted least-norm(WLN) solution. In the new approach, the original advantage of avoiding the joint limit to get the optimized solutions in WLN is saved and the new advantage of improving the accuracy of inverse kinematics solution is developed. Moreover, one open problem in this field that Jacobi pseudo-inverse does not exist in some special situations can be solved properly based on the iteration algorithm. Simulation results clearly show that the proposed approach can achieve better tracking accuracy of desired trajectories and solve the open problem mentioned above.