Adaptive neural learning control of rigid-link electrically-driven robot manipulators

Based on deterministic learning theory which states that an appropriately designed adaptive neural controller can learn the unknown system internal dynamics during a stable control process, this paper investigates deterministic learning from adaptive neural control of rigid-link electrically-driven (RLED) robot manipulators with completely unknown system dynamics. Firstly, the recent results on localized RBF networks and stability analysis of linear time-varying (LTV) systems are presented. Secondly, a stable adaptive neural control algorithm is designed for RLED robot manipulators, and the closed-loop control system with the LTV form is obtained. Deterministic learning of RLED robot manipulators is analyzed, locally-accurate approximation of the closed-loop control system dynamics is achieved along the periodic tracking orbit. Improved control performance is achieved using learned knowledge stored as a set of constant neural weights. Finally, simulation example is presented to demonstrate the effectiveness of the proposed control algorithm.