Model Predictive Control of Autonomous Surface Vehicles for Obstacle Avoidance Using High-Order Control Barrier Functions

Obstacle avoidance is a fundamental requirement for autonomous surface vehicles (ASVs). However, in complex marine environments, multimodal disturbances and the inherent constraints of ASV s further increase the difficulty of obstacle avoidance. Additionally, diverse obstacle types pose challenges to tracking safety. This research presents a model predictive control (MPC) framework utilizing high-order control barrier functions (HOCBFs) to ensure safe and stable obstacle avoidance for ASVs. The framework integrates HOCBFs directly as constraints into the optimization problem of control barrier functions, thereby guaranteeing safety while addressing system constraints. Exper-imental results show that the proposed approach outperforms existing methods in simulation-based tests.