A Single Network Adaptive Critic-Based Redundancy Resolution Scheme for Robot Manipulators

摘要

This paper proposes an adaptive critic-based real-time redundancy resolution scheme for kinematic control of a redundant manipulator. The kinematic control of the redundant manipulator is formulated as a discrete-time input affine system, and then, an optimal real-time redundancy resolution scheme is proposed. The optimal control law is derived in real time using an adaptive critic framework. The proposed single network adaptive critic-based methodology defines the additional task in terms of integral cost function which results in global optimal solution. With adaptive critic-based optimal control scheme, the optimal redundancy resolution is achieved in real time without the computation of inverse, which makes the method computationally efficient. The problem formulation as proposed in this paper is first of its kinds. Furthermore, the real-time optimal redundancy resolution using an integral cost function has been comprehensively solved, which is also a novel contribution. The proposed scheme is tested in both simulation and experiment on a seven-degree-of-freedom PowerCube manipulator from Amtec Robotics.