A Multi-Stage Heading Control Approach for Autonomous Usv Navigation in Gnss-Denied Environments with Uav Cooperation

Unmanned Surface Vessels (USVs) require robust navigation strategies for tasks like reaching specific targets autonomously in adverse weather conditions. Typically, the target position is well-defined, and USV localization is performed using Global Navigation Satellite Systems (GNSS) sensors. However, GNSS-denied environments, uncertain target locations, and harsh weather conditions make it difficult for traditional navigation approaches to work properly. This paper presents a novel control-based framework for robust autonomous navigation of a USV in GNSS-denied maritime environments with uncertain target locations. The framework employs a heterogeneous robotic system consisting of an Unmanned Aerial Vehicle (UAV) and a USV. It uses a multi-stage heading control approach with autonomous mode switching based on target proximity. Initially, UAV provides localization and guidance for the USV. As the target enters the USV's camera field-of-view, navigation modes switch to visual servoing. Finally, upon target acquisition within the LiDARs range, the LiDAR-based navigation stage recalculates the precise position of the target vessel to minimize the uncertainty of the target pose for precise maneuvering.The framework's effectiveness was validated in real-world sea conditions during the Muhammad Bin Zayed International Robotic Competition. The results demonstrate the proposed framework's potential for robust USV navigation in extreme weather conditions.