Adaptive PD networks tracking control with full-state constraints for redundant parallel manipulators

In this paper, a detailed study to apply an adaptive proportional-derivative networks (APDNs) tracking control is proposed for redundant parallel manipulators with full-state constraints. APDNs is designed - based on the combination of the nonlinear compensation, the PD nonlinear sliding part, and feedforward radial basic function neural networks (RBFNs) with online learning. By using the Lyapunov method, the stability of the closed-loop system in the case of full-state constraints is proven. The proposed control strategy has better dynamic performance and higher robustness in comparison with PD-sliding mode control and augmented nonlinear PD control. The effectiveness proposed control is illustrated by simulation.