Adaptive-observer-based formation tracking of networked uncertain underactuated surface vessels with connectivity preservation and collision avoidance

This paper investigates an adaptive output-feedback formation tracking problem for ensuring connectivity preservation and collision avoidance among networked uncertain underactuated surface vessels (USVs) with different communication ranges. An adaptive observer using neural networks is designed to estimate the velocity information of USVs where neural networks estimate unknown nonlinearities of USVs. Especially, contrary to the existing related work of USVs, a new state transformation technique for the adaptive observer design is presented to relax the condition requiring the boundedness of the yaw velocity of USVs. Then, the recursive tracker design strategy is established by using a unified error function for connectivity-preserving and collision-avoiding formation tracking, without employing any potential functions. The proposed formation tracker does not require additional neural networks to estimate unknown nonlinearities derived from the tracker design procedure. The proposed theoretical result is proved in the sense of Lyapunov. (C) 2019 Published by Elsevier Ltd on behalf of The Franklin Institute.