Mapping, Path Optimization, and Motion Control of a Robotic Seawater Waste-Management System

This paper presents a novel robotic system designed to perform the task of waste removal from seawater. The system comprises an unmanned aerial vehicle (UAV) for surveying the area to be cleaned combined with an unmanned surface vehicle (USV) for waste retrieval. Initially, the paper highlights the use of a color segmentation algorithm alongside a Kalman filter for waste detection from the UAV’s perspective, and for tracking a non-stationary USV within the UAV’s field of view. Subsequently, the paper addresses the challenge of waste detection from the USV’s viewpoint using the Hough Transform. The extracted waste coordinates from the detection algorithm are fed into an ant colony optimization (ACO) path-planning algorithm. Furthermore, a dynamic model for the USV is derived, and a control law is proposed to guide the USV along the desired path. Finally, the paper discusses simulation results highlighting a 0.6 Figure of Merit (FOM) score for waste detection and relatively accurate path-following of the USV.