MPC-CBF Strategy for Multi-Agent System Obstacle Avoidance Path-Following

This paper presents a novel safety-critical distributed model predictive control method for multi-agent systems based on Control Barrier Functions (CBFs). Unlike traditional MPC approaches that rely on terminal constraints, a distributed Lyapunov-based model predictive control (DMPC) framework is proposed, which effectively decouples the optimization of system stability and performance. Within the DMPC framework, safety is guaranteed through the use of Control Barrier Function (CBF) constraints, leveraging the rolling horizon optimization of MPC to proactively avoid collisions. To reconcile the potential conflict between the stability and safety objectives of the MPC algorithm, slack variables are introduced into the CBF constraints. Finally, simulation studies on path following are conducted to validate the effectiveness of the proposed safety-critical robust control scheme.