Toward Unified Adaptive Teleoperation Based on Damping ZNN for Robot Manipulators With Unknown Kinematics

Abstract

Traditional teleoperation systems are usually designed for a single class of slave robots and require the specific kinematic model. This leads to the problems of high computational cost, poor robustness, and poor portability in traditional teleoperation systems. In this article, a unified adaptive teleoperation system is proposed for different kinds of slave robots. To this end, a closed-loop control system based on damping zeroing neural network (DampZNN) is proposed to achieve the unified kinematic control of both the redundant robot and the continuum robot with unknown kinematics. The control system is constructed based on two DampZNNs, which are utilized to estimate the unknown kinematic model and to solve the inverse kinematics problem respectively. By combining the DampZNN control system with teleoperation technology, a unified adaptive teleoperation system is proposed for human operators to teleoperate both the redundant slave robot and the continuum slave robot easily. The efficacy of the proposed teleoperation system is verified by experiments based on different slave robots and the advantages of the proposed teleoperation system are revealed by comparative studies.