Barrier-Certified Model Predictive Cooperative Path Following Control of Connected Autonomous Surface Vehicles

Abstract

This paper investigates the safe cooperative path following of connected autonomous surface vehicles subject to static and dynamic obstacles, as well as physical constraints. A barrier-certified model predictive cooperative path following control method is proposed with the capability of collision avoidance and constraint satisfaction. Specifically, a robust-exact-differentiators-based extended state observer is employed to estimate the unknown kinetics including model uncertainties and external disturbances. Based on a path variable containment approach, a nominal receding-horizon control law is designed to achieve cooperative path following task within the physical constraints. A safe optimal control law is designed based on control barrier functions to generate optimal surge force and heading angle within the safety constraints. A receding-horizon heading control law is designed to track the desired heading signals. Constrained quadratic programming problems are formulated and solved via neurodynamic optimization. Simulation results are elaborated to validate the efficacy of the proposed barrier-certified model predictive control method for cooperative path following subject to static and dynamic obstacles.