MPC-Based Motion Planning and Tracking Control for Autonomous Underwater Vehicles

This paper proposes a motion planning and tracking control method based on model predictive control (MPC) for autonomous underwater vehicles (AUVs). Traditional planning algorithms do not smoothly search for paths on grid maps and fail to satisfy AUV’s kinematic constraints. To address this, a new objective function with an obstacle avoidance function is designed to reduce excessive avoidance in traditional algorithms. The motion planning controller is designed to find a kinematically feasible obstacle avoidance trajectory by solving a nonlinear optimization problem with constraints based on AUV state information, obstacle information, and reference target information. The MPC trajectory tracking controller combines the planned trajectory to track it. Simulation results validate the effectiveness of the proposed method.