Density fronts and submesoscale filaments are prominent features in coastal and open oceans. Here, dissolved contaminants and other pollutants, such plastic debris, may accumulate and pose hazards to the marine environment. Studying these fronts and the accumulation processes in detail has proven difficult with standard oceanographic methods, mainly because research vessels and other measurements platforms have a strong influence on the structures themselves. In the context of the project Saimidris, a novel approach will be developed that combines knowledge from the fields of engineering, applied oceanography and artificial intelligence. The aim is to develop a smart swarm of autonomously acting drifters equipped with miniaturized pollutant sensors to obtain a detailed view of the temporal and spatial dynamics under various environmental conditions. A specific feature will be the propulsion of the drifters that will allow them to move actively with rigid sails or alternative techniques. Digital ad hoc networks and AI technologies will enable the drifters to sense the environment and adapt their sampling strategy to the ambient conditions. The collaboration will enhance the digitalization of ocean observation systems used for assessing anthropogenic hazards under highly variable environmental conditions.