Marine Oil-Spill Confinement and Cleanup with Boom-Towing Vessel Fleets

Marine oil spills damage marine ecosystems, con- taminate coastlines, and disrupt food webs, while imposing heavy economic losses on fisheries, tourism, and coastal communities. Prior work has demonstrated the potential to contain and clean individual spills using a duo of autonomous surface vehicles (ASVs) equipped with a towed boom and skimmers. Existing al-gorithmic methods mostly focus on isolated slicks and individual ASV duos, and lack scalable approaches for routing large robotic fleets to rapidly address such environmental disasters while capturing realistic problem aspects. In this work, we propose an effective multi-robot approach for confinement and cleanup using coordinated ASV duos. We model this problem as a risk- weighted, minimum-latency problem that prioritizes spills based on their volume, proximity to sensitive assets, and the time it takes for the ASV duos to reach and clean them. We then develop an effective solution to the above problem based on mixed-integer linear programming that can tackle realistic problem instances within several minutes on a commodity laptop. Finally, for real-world execution, we develop and evaluate two controllers—one based on feedback linearization with proven stability, and another on the popular PID approach—for accurate path tracking of a boom-carrying ASV duo. Together, these contributions provide a scalable, computationally efficient, and practically deployable framework for rapid, risk-aware multi-robot response to large-scale marine oil spills.

M.Sc. student under the supervision of Prof. Kiril Solovey.