Adaptive operation-space control of redundant manipulators with joint limits avoidance

In this paper, adaptive operation-space control with joint limits avoidance is proposed for a redundant robot manipulator. Utilizing redundant properties of the robotic manipulator, joint limits avoidance is achieved without interfering the main-task objective in the operation-space. Two control objectives are unified under one common control framework. To cope with the unknown and unstructured dynamic nonlinearities of the robot model, neural networks (NNs) are constructed under the proposed torque control design as to approximate the unknown dynamics and achieve small tracking errors. Simulation studies are carried out to verify the effectiveness of the proposed framework.