PhD Presentation: A hyper-heuristic approach to achieving long-term autonomy in a heterogeneous swarm of marine robots

A heterogeneous swarm of marine robots was developed with the goal of autonomous long-term monitoring of environmental phenomena in the highly relevant ecosystem of Venice, Italy. For the purpose of making long-term autonomy possible, task allocation and sequencing are introduced into the system’s energy management and energy sharing procedures. In a scenario where the system needs to autonomously go about its monitoring mission and survive for an extended amount of time, the available maximum capacity of 5 USVs – aPad platforms which represent the charging hubs of the system – is usually outnumbered by the number of active charging requests, leading to a need for careful planning and optimisation of robot activities. Thus, a two-layered system of decision-making algorithms is developed: a low-level specific solution-focused set of algorithms, and a high-level hyper-heuristic which selects between them depending on various performance indices. As logistics are a continuing challenge in the field of marine robotics, especially when dealing with a large number of agents to be collected and redeployed per experimental run, a testing approach is needed that provides the benefits of simulation while also reflecting the complexity of the real world. The development of a vehicle-in-the-loop test environment in which a surface station simulates and transmits the data of any number of simulated agents, while a real marine platform operates based on the received information, is also described. Several experimental runs of a specific use-case test scenario using the developed framework and carried out in the field are described and their results are examined.