A novel neuro-based model reference adaptive control for a two link robot arm

This paper presents a novel neuro-adaptive control for a two link robot arm. The learning algorithm guarantees the stability of a class of closed loop neural network control systems. The underlying control system, here a robot arm, represents a nonlinear system. The neuro-controller, which indeed represents a direct adaptive controller, guarantees the closed loop stability for any arbitrary initial values of states, neural network parameters and any unknown-but-bounded disturbances, provided that some soft conditions are satisfied. No additional controllers or robustifying terms are needed. Neural network weight matrices are adapted online with no initial offline training. To show the ability of the proposed neuro-controller on the robot arm system, extensive simulations have been performed. The simulation results better judge the merit of the neuro-adaptive control scheme