MLP-Based Fixed-Time Neural Network Formation Control for Uncertain Multi-USVs with Nonsymmetric Dead Zone

This paper investigates the fixed-time formation control of uncertain multiple unmanned surface vehicle systems with nonsymmetric dead zones and environmental disturbances. A fixed-time neural network sliding mode formation control scheme (FTNNFC) is developed for the formation control of the multi-USVs. In addition, a minimum-learning-parameter (MLP)based neural network is introduced to estimate the uncertain nonlinear dynamics and time-varying environmental disturbances. Theoretical analysis demonstrates that under the proposed scheme, the multi-USVs achieves fixed-time stability, and the formation errors are converged to a bounded region within a fixed time, regardless of the initial conditions. At last, numerical simulations are conducted to validate the effectiveness of the developed strategy.