Collision-Free Cooperative Path Following of USVs Based on Concurrent Learning Switching Extended State Observers

This article addresses the collision-free cooperative path following (CPF) problem of multiple underactuated unmanned surface vehicles (USVs) in the presence of stationary and dynamic obstacles. The position and velocity of the obstacles are unknown, and only the relative angle and relative distance are measured. A collision-free control method is proposed for achieving collision avoidance and CPF for USVs. Specifically, at the kinematic level, collision-free guidance laws are proposed for CPF based on time-to-collision feedback and fixed-time coordination. Both stationary and dynamic obstacles can be avoided by incorporating time-to-collision feedback into the desired yaw rate. A fixed-time path variable update law is applied to synchronize the path variables in fixed time. At the dynamic level, concurrent learning switching extended state observers are proposed for estimating the model uncertainties including the unknown control input. As a result, the proposed data-driven kinetic controller does not need any parameter information. The hardware-in-loop simulation and experiment results demonstrate the effectiveness of the proposed collision-free CPF controller for USVs.