抄録
This paper presents Inverse Kinematics algorithms using Nonlinear Least-squares Method for Redundant manipulator. Newton-Raphson method (NR) is a common method to calculate Inverse Kinematics. However, numerical solution using NR method produces problems on singularity and solvability. If numerical solution using NR becomes unstable, robot could be out of control. Hence, it requires tedious check on the stability, prior to the robot operation. Therefore, this paper proposes a robust numerical solution using Levenberg-Marquardt (LM) method. The method has strong stability against the singularity and solvability. In addition, Weighted Least-Norm method (WLN) is used to keep the solution inside the movable range of the robot. Finally, we confirm the validity of the proposed method through the simulation results.
