Wild Swarms: Autonomous Drones for Environmental Monitoring and Protection

In this paper, we present an example case study of how an autonomous swarm-based Multi-Robot System (MRS) could contribute to environmental protection. Our proof-of-concept simulation involves using a collective intelligence framework to manage a drone swarm tasked with protecting a wilderness area over a long period of time. In the proposed ambitious and futuristic Concept of Operation (ConOps), a self-sufficient and self-organizing colony of “drone wardens” is left to autonomously plan and execute thousands of individual reconnaissance missions, attempting to maintain the best possible situational awareness while simultaneously managing its resources and responding to relevant actors and events in the environment. Specifically, we have devised a scenario in which two types of external actors (hikers (blue team) and bears (red team)) should elicit a different collective response from the swarm: whilst the former should only be checked on from time to time to ensure their safety, the latter should be monitored continuously during their crossing of the park. Numerical experiments, in which a variety of simple but robust decision-making algorithms were tested, demonstrate our system’s ability to support the relevant swarming behaviour, such as efficient division of labour and recruitment, in the pursuit of this mission.