MRAC for nonlinear systems with unknown gain signs and unmodeled dynamics

Abstract

Based on the property of Nussbaum function and the approximation capability of neural networks, model reference adaptive control (MRAC) is presented for a class of nonlinear systems with unknown gain signs and unmodeled dynamics. The approach eliminates the requirement for a priori knowledge of the control gain sign to be known by using the property of Nussbaum function. A dynamic signal is introduced to deal with the dynamic uncertainty problem. It is proved that all the signals in the closed-loop control system are semi-globally uniformly ultimately bounded. Simulation results demonstrate the effectiveness of the proposed approach.