Robust adaptive single neural control for a class of uncertain nonlinear systems with input nonlinearity

This paper presents an adaptive single neural controller for a class of uncertain nonlinear systems subject to a nonlinear input. A new type of neuron called auto-tuning neuron with three adjustable parameters will be introduced to construct a single neural controller. From the concept of the sliding mode control, a simple adaptation law, minimizing the value of a designed sliding condition based on a modified MIT rule, is developed for online updating these parameters in the auto-tuning neuron, even if the nonlinear plant considered is with the uncertainty, external noisy perturbation, and nonlinear input. Lastly, a controlled well-known Duffing Holmes chaotic system is illustrated to show the effectiveness of the proposed neural controller.