Numerical method for inverse kinematics using an extended angle-axis vector to avoid deadlock caused by joint limits

In this paper, a new inverse kinematics strategy is proposed for avoiding deadlock caused by joint limits. The proposed method is a combination of the conventional numerical method for inverse kinematics and an extended angle-axis vector. The extended angle-axis vector, instead of Euler angles, represents the orientation error. In the proposed method, the rotation direction of the end effector is changed by using the extended angle-axis vector to avoid deadlock. This paper describes the specific parameters included in the extended angle-axis vector that are necessary to change the rotation direction. The effectiveness of the proposed method was verified through simulations using a 9-degree-of-freedom robot. The proposed method is expected to be useful for solving inverse kinematics problems, which remain analytically unresolved.