Barrier function-based adaptive fixed-time ADLESN prescribed performance tracking control for UUVs with input saturation

This work addresses the adaptive fixed-time trajectory tracking problem for an unmanned underwater vehicle (UUV) under asymmetric input saturation, modeling uncertainty, ocean current disturbances, and output constraints. A novel fixed-time performance function is designed, and a control framework based on an asymmetric barrier function and prescribed performance control (PPC) is constructed. The tracking performance is ensured based on this framework, and the error caused by the error transformation necessary for the PPC strategy is also considered. Then, a practical fixed-time theorem is introduced, based on which a fixed-time adaptive filter is designed to avoid the complexity explosion problem. Then, an adaptive double-layer echo state network (ADLESN) that enhances the sparsity of the output weights is proposed, and the parameters of the activation function of the network can be adjusted adaptively. The ADLESN is employed to estimate the upper bounds on the system concentration of uncertainty, and an adaptive error compensation system is designed for the network to improve the approximation accuracy. Furthermore, the convergence of all errors in fixed time is verified by employing Lyapunov stability analysis, and the potential singularity problem is avoided. Finally, comparative experiments based on numerical simulations are presented to demonstrate the benefits of the algorithm. © 2024 Elsevier Ltd