Multi-Objective Nonlinear Model Predictive Control for Tethered USV-ROV Cooperative Tracking and Dynamic Obstacle Avoidance

Tethered unmanned surface vehicle (USV) and remotely operated vehicle (ROV) systems are widely used in deep-sea inspection, observation, and intervention tasks. During cooperative operations, the USV must follow the mission trajectory of the ROV while avoiding surface obstacles and maintaining a prescribed tether-related safety envelope. This study proposes a multi-objective nonlinear model predictive control (NMPC) framework for USV-side cooperative tracking and dynamic obstacle avoidance in tethered USV-ROV operations. The framework integrates the predicted three-dimensional ROV trajectory, USV nonholonomic motion, surface-obstacle avoidance, straight-line tether-length-related geometric constraints, and control-smoothness regulation into a unified receding-horizon optimization problem. Sequential Least Squares Programming is used to compute the online control sequence. Numerical simulations include obstacle-free tracking under bounded ocean-current disturbances and heterogeneous surface–underwater obstacle scenarios. The results show that the proposed controller provides improved tracking performance and maintains the straight-line tether-length proxy below the prescribed limit in the tested simulations. The current-disturbance results further indicate preliminary disturbance-rejection capability under bounded time-varying ocean currents. The study provides a controller-level numerical framework for cooperative tracking and surface obstacle avoidance in tethered USV-ROV operations.