AI-Driven Marine Waste Towing System for Oceanic Plastic Mitigation in East and West Africa

Marine plastic pollution constitutes a critical and intensifying threat to the coastal ecosystems, fisheries, and blue economies of East and West Africa, yet prevailing mitigation responses remain reactive, labour-intensive, and poorly matched to the region's infrastructural and governance realities. Existing autonomous solutions, including in-port collection vessels such as ClearBot and WasteShark and passive offshore barriers, are constrained to narrow operational envelopes and do not provide integrated detection, consolidation, and redirection at coastal scale. This paper addresses that gap by advancing the Marine Waste Towing System (MWTS) — a conceptual framework in which a networked fleet of autonomous, AI-enabled surface vessels detects, tracks, and tows floating plastic toward designated recovery zones under cloud-based coordination. The framework is grounded in systems theory, cyber-physical and autonomous-systems perspectives, sustainability-transitions thinking, and blue-economy policy, and is developed through a systems-engineering methodology comprising functional decomposition, architecture modelling, operational-workflow modelling, and scenario-based analysis. We introduce simplified operational models for detection efficiency, patrol coverage, route optimisation, towing and consolidation, fleet coordination, energy endurance, and overall recovery effectiveness, and present a multi-criteria comparison of the MWTS against incumbent technologies. The analysis is balanced by a critical appraisal of technological, ecological, cybersecurity, economic, and regulatory constraints — including alignment with the newly adopted IMO MASS Code, the Nairobi and Abidjan Conventions, and African Union Agenda 2063. The MWTS is offered not as a finished prototype but as a rigorous, regionally grounded design framework intended to guide research, investment, and policy toward scalable marine-waste management in the Global South.