Review of the state of the art in control laws for marine robots

Abstract

This report presents a comprehensive review of the state of the art in control and navigation strategies for Unmanned Surface Vehicles (USVs), with a focus on underactuated platforms operating in uncertain marine environments. It first explores the evolution, classification, and architecture of USVs, emphasizing their growing role in environmental monitoring, inspection, and surveillance missions. The study also examines the ROS 2 software framework, highlighting its modularity, scalability, and suitability for distributed robotic systems. A detailed analysis of the main sensor technologies including IMU, DVL, GNSS, magnetometers, and acous- tic positioning systems is provided, along with modern data fusion techniques such as the Extended and Unscented Kalman Filters (EKF, UKF) and Complementary Filters. The second part of the report reviews control approaches used for underactuated marine robots, ranging from classical PID to advanced methods like Model Predictive Control (MPC), Backstepping, Sliding Mode Control (SMC), and Adaptive Control. Comparative evaluations underline the trade-offs between simplicity, robustness, and computational com- plexity, showing that hybrid architectures combining geometric guidance (LOS, Pure Pursuit, or Vector Fields) with robust or adaptive control provide the best performance for trajectory tracking under disturbances. This literature review establishes a theoretical foundation for future implementation and optimization of control systems for low-cost autonomous surface vehicles.