Adaptive control of non-affine nonlinear systems using neural networks

An adaptive control approach is presented for a class of nonaffine nonlinear systems using neural networks and sliding mode control. Firstly, it is shown that the control of nonaffine nonlinear system is equivalent to the control of the affine nonlinear system in the neighborhood of the operating trajectory under mild conditions. Then, an adaptive neural sliding mode controller is designed for the transformed nonlinear system. The proposed neural sliding mode controller can guarantee that the output tracking error of the system asymptotically converges to zero. Simulation results illustrate the effectiveness of the scheme.