Online neural network compensator for uncertain nonlinear MIMO systems based on RISE feedback

The paper proposes an adaptive controller to track time-varying trajectory for nonlinear MIIMO systems in presence of uncertainty and disturbances. The suggested controller employs an online adaptive Gaussian Radial Basis Function (RBF) network to estimate nonlinear functions together with Robust Integral of the Sign of the Error (RISE) control strategy. RISE term is injected in control structure in order to eliminate neural network error and external disturbances. The weight matrix and Gaussian functions update laws of adaptive RBF that are obtained of Lyapunov analysis tuned based on state error information in online manner. Hence, there are no need to choice of centers and width of Gaussian function that are essential for RBF desirable acting. The closed-loop stability is guaranteed by a Lyapunov stability analysis. Finally, simulations on two link robot manipulator illustrate the performance of the derived tracking control technique

Online neural network compensator for uncertain nonlinear MIMO systems based on RISE feedback Keywords:

Adaptive RBF neural network , Lyapunov analysis , Robust Integral of the Sign of the Error (RISE) , uncertainty