Fixed-time adaptive neural network compensation control for uncertain nonlinear systems

Uncertainties are the main obstacle to improving the control performance of nonlinear systems. To address this challenge, this paper proposes a fixed-time adaptive neural network compensation control method for a class of high-order nonlinear systems exhibiting both uncertain nonlinearities and parametric uncertainties. Specifically, a fixed-time adaptive neural network (FTANN) is designed to deal with uncertain nonlinearities, while a new fixed-time adaptive law is developed to address parametric uncertainties. These two approaches are integrated with a gain-adaptive fixed-time filter within the dynamic surface control (DSC) framework. This not only resolves the "differential explosion" problem but also reduces the conservatism of the controller by lowering the robust feedback gain. As proven through Lyapunov analysis, the proposed controller guarantees fixed-time stability for all system states. Comparative simulations and experimental results provide further evidence supporting the effectiveness of the proposed controller.