Event-triggered time-synchronized based neural network formation control of heterogeneous USV-AUV systems with locked-in-place actuators

This paper investigates time-synchronized formation control for heterogeneous systems consisting of unmanned surface vessels (USVs) and autonomous underwater vehicles (AUVs) in the presence of actuator lock-in-place faults. A neural network-based dynamic event-triggered fixed-time synchronized sliding mode control (NNDET-FTSSMC) approach is proposed to address both time-invariant and time-varying formation tracking problems under system uncertainties. The neural network is employed to approximate and compensate for unknown nonlinear dynamics , thereby reducing the dependence on precise prior knowledge of system uncertainties and disturbances while ensuring computational efficiency for resource-constrained maritime systems. To further alleviate the communication burden, an adaptive distributed event-triggered mechanism is developed to dynamically adjust the triggering threshold. Theoretical analysis demonstrates that maintaining ratio persistence between the formation tracking errors and the sliding mode surface guarantees that all state variables achieve synchronized convergence within a fixed time, while effectively avoiding the Zeno phenomenon. Finally,the numerical simulations are conducted to verify the effectiveness and advantages of the method.