Evolving Robust Supervisors for Robot Swarms in Uncertain Complex Environments

Abstract

Whilst swarms have potential in a range of applications, in practical real-world situations, we need easy ways to supervise and change the behaviour of swarms to promote robust performance. In this paper, we design artificial supervision of swarms to enable an agent to interact with a swarm of robots and command it to efficiently search complex partially known environments. This is implemented through artificial evolution of human readable behaviour trees which represent supervisory strategies. In search and rescue (SAR) problems, considering uncertainty is crucial to achieve reliable performance. Therefore, we task supervisors to explore two complex environments subject to varying blockages which greatly hinder accessibility. We demonstrate the improved performance achieved with the evolved supervisors and produce robust search solutions which adapt to the uncertain conditions.

The Impact of Network Connectivity on Collective Learning

“…In swarm systems, the ability for agents to make decisions in a distributed manner is crucial for their real-world deployment, as robot swarms are appropriate platforms for deployment in difficult terrain and hazardous environments, such as for the detection of wildfires [12]. Due to their large scale, swarm systems often cannot be supervised at the individual level and must instead rely on human supervision at the swarm level [11], while making its own decisions and taking actions autonomously, often by reaching a consensus [19] or through quorum sensing [1]. Much of the literature on collective decision-making in robot swarms centres on the best-of-n problem; a bio-inspired problem in which the swarm must decide which is the best option of n possible alternatives [15].…”