For the design and verification of a motion control algorithm for unmanned surface vehicles, a simulation system is developed based on VRX and Simulink. Firstly, considering the effect of wind, a dynamic model of the USV with podded propellers is established. Secondly, combined with speed control, three control modes are considered, including yaw rate control, heading control, and path-following control, and speed, heading, yaw rate, and path guidance controllers are designed. Then, a real-time simulation system is developed based on the Virtual RobotX (VRX) environment and the Simulink ROS2 toolbox. Finally, motion control simulation experiments under three control modes and a path-following water tank experiment are carried out. The designed simulation system can simulate the motion of USVs and different environmental elements, such as wind, intuitively and realistically. In simulation experiments, the designed controllers can make the USV follow commands quickly and accurately under three control modes. In the water tank experiment, the USV could stably track the desired path with a relatively small tracking error. Therefore, the effectiveness of the simulation system is strongly confirmed through simulation experiments and the water tank experiment. The simulation system will be expanded in the future for more research on target recognition, path planning, and other aspects of USVs.

To address the control coupling challenges arising from task heterogeneity of unmanned surface vehicle (USV) formation, a distributed hybrid deep reinforcement learning (HDRL) framework is proposed. The framework decomposes the formation task into two subtasks: leader path planning using the single-agent deep reinforcement learning (SADRL) algorithm and follower formation tracking using the multi-agent deep reinforcement learning (MADRL) algorithm. By embedding the physical constraints of the real Otter USV into the training loop, the policy network outputs are mapped to propeller revolutions that conform to its dynamic characteristics. To optimize control performance, a dynamic gating mechanism triggered by formation position error is developed to mitigate multi-objective interference through temporal task scheduling. Concurrently, a mirror mapping mechanism leveraging the physical symmetry of the formation is designed to achieve policy sharing and data augmentation. Furthermore, the desired velocity calculated based on rigid-body kinematics is used to achieve kinematic-compensated formation tracking. The simulation results indicate that, compared to the SADRL algorithm, the planning success rate of HDRL is improved by 44.59%. Furthermore, compared to the MADRL algorithm, the integrated tracking performance is enhanced by 21.79–39.64%.