A Milling Swarm of Ground Robots using Spiking Neural Networks

Spiking Neural Networks, or SNNs, have the potential to enable ultra-cheap, ultra-small neural controllers for robots, as the bio-plausibility of SNNs implies the possibility of microscopic, bio-scale control circuits. We seek to explore the viability of using SNNs to form agent-local policies on robots with only a simple 1-bit visibility sensor. Using simulations, we evolve a compact (14 neurons) SNN which controls robots to move in a circular milling formation. Based on whether a robot can see any peers in its field of view, the SNN decides to move either forwards or backwards while turning either left or right. An evolutionary process evolves not only the weights, thresholds, and synaptic delay parameters, but the connectivity, structure, and size of the network as well in order to optimize a global Circliness metric describing the milling swarm formation. We transfer this behavior to a swarm of ground vehicles with no wheel encoders or other localization capabilities. To enable this sim2real transfer, the simulated agent is matched to the robotic embodiment through an iterative process of characterization, the RSRS process. The resulting emergent behavior is decentralized and robust, requires little to no perception, localization, or planning, and achieves comparable milling performance to existing controllers. Yet, the learned control policy is distinct from that of the state-of-the-art. We hope this motivates further exploration of the use of SNNs in swarms of robots.