Adaptive neural quantized formation control for heterogeneous underactuated ships with the MVS guidance

This paper presents a novel neural formation controller for the heterogeneous underactuated surface vehicle (USV) with quantized input signal. Within the leader–follower framework, the logical virtual ship and guidance virtual ship are introduced to facilitate the smooth tracking control of real ship, which is called the multi-virtual ships (MVS) guidance principle. By fusion of the dynamic surface control and neural networks, the analytical derivative of virtual controller is avoided and the unknown dynamics is approximated. To reduce the unduly execution of actuators, the control input is modified by the hysteresis quantization technique. Meanwhile, two different adaptive parameters are designed to deal with the model inconsistency caused by heterogeneous ships and the signal transition caused by hysteresis quantizer respectively. On the basis of Lyapunov analysis, all the signals of the closed-loop control system are guaranteed to be semi-global uniform ultimate bounded (SGUUB). Under the scenario of ship formation ply across a narrow strait, numerical simulations are presented to verify the proposed algorithm. •The MVS guidance is designed for heterogeneous ships to generate the reference.•The hysteresis quantizer is applied to reduce the execution frequency of actuators.•Adaptive parameters are derived to stabilize instability terms of heterogeneous ship.