Abstract: For the trajectory tracking control of USV (unmanned surface vehicle) swarms, we propose a collision avoidance tracking control system based on obstacle-avoidance heading reference angle, considering model uncertainties, environmental disturbances, input saturation, and other factors, to address the problems of collision avoidance difficulty and slow tracking in swarm control of unmanned boats. To solve the collision avoidance problem, we design an obstacle-avoidance heading reference angle based on the relative distance and angle between the unmanned boat and the obstacle. To solve the slow tracking problem, we redefine the relative distance between unmanned boats based on the artificial potential function. We design the swarm controller based on the artificial potential function and the obstacle-avoidance heading reference angle. The stability of the closed-loop control system is proved by the Lyapunov direct method. The performance of the proposed control method is verified through experimental simulations. The experimental results show that the proposed method achieves faster convergence speed for collision avoidance and trajectory tracking control by utilizing artificial potential functions and the obstacle avoidance heading reference angle. Additionally, the proposed controller effectively balances the trade-off between collision avoidance and input saturation, enabling smooth obstacle avoidance.