Fully Distributed Target Encircling Control of Autonomous Surface Vehicles Based on Noncooperative Games

Abstract

This paper addresses cooperative target encircling of multiple autonomous surface vehicles (ASVs) with private and potentially competitive objectives. A fully distributed encircling control approach is proposed based on noncooperative games. Specifically, a fully distributed estimator with an adaptive gain is developed to estimate the target information without using global state or topology knowledge. Based on a low-frequency learning technique, a fuzzy predictor is presented to approximate the unknown vehicle kinematics induced by uncertain nonlinearities and environmental disturbances. By decoupling the cooperative target encircling into an encircling task and a spacing task, an encircling control law and a spacing control law are designed based on fully distributed Nash equilibrium seeking for achieving the private control objective of each ASV. The input-to-state stability of the closed-loop system is proven via cascade analysis. Simulation results are provided to illustrate the effectiveness of the noncooperative game-based control method for ASVs in circumnavigation missions.