Adaptive neural networks control for MIMO nonlinear systems with unmeasured states and unmodeled dynamics

This paper presents an adaptive backstepping control scheme based NNs technique for a class of un- certain nonlinear MIMO systems with unmeasured states and unmodeled dynamics. The contributions of this paper are listed as follows:•The unmodeled dynamics of the system are processed by a dynamic signal. Then, the combinational uncertainties caused by unmodeled dynamics, unknown nonlinear functions and dynamic disturbances are approximated by neural networks.•An observer is established to estimate the unmeasured states in the system. It is proved that the observer error vectors can be made arbitrarily small by selecting appropriate parameters. In this paper, an adaptive neural networks control scheme is developed for a class of multi input and multi output uncertain nonlinear systems with unmeasured states and unmodeled dynamics. In the control scheme, a dynamic signal is used to deal with the unmodeled dynamics and a neural observer is designed to estimate the unmeasured states. Meanwhile, the neural networks are used to estimate the combinational unknown nonlinear function at each step of backstepping process. It is proved that all signals of the closed-loop system are semi global uniformly ultimately bounded (SGUUB). Finally, a simulation example is provided to show the effectiveness of the proposed control method.