Robust Zhang Neural Network for Tracking Control of Parallel Robot Manipulators With Unknown Parameters

Abstract

Under the situation of parameter uncertainty, the tracking control of parallel robot manipulators is a challenging problem in robotic research. Unlike conventional Zhang neural network (ZNN) relying on the assumption that the robot parameter information is fully and accurately known, this paper proposes a robust Zhang neural network (RZNN) for tracking control problems solving of parallel robot manipulators in the absence of parameter information. The proposed RZNN features the full utilization of effector feedback information, and shows a robust tracking performance even with unknown robot parameter information. Then, the continuous-time model of the RZNN is discretized via Euler forward formula (EFF) for numerical implementation. Finally, comprehensive simulative experiments including robustness test verify the effectiveness of the RZNN model for the real-time tracking control of parallel robot manipulators with unknown parameters.