Discovery and Deployment of Emergent Robot Swarm Behaviors via Representation Learning and Real2Sim2Real Transfer
This work addresses the challenge of behavior discovery for robot swarms, enabling automated deployment without human intervention, though it builds incrementally on existing sim2real transfer techniques.
The paper tackles the problem of automatically discovering emergent behaviors for robot swarms by introducing a method that combines representation learning and novelty search in simulation, enabling direct transfer to real robots. It shows that this approach outperforms previous hand-crafted metrics in simulation and successfully deploys discovered behaviors on a low-cost robot platform.
Given a swarm of limited-capability robots, we seek to automatically discover the set of possible emergent behaviors. Prior approaches to behavior discovery rely on human feedback or hand-crafted behavior metrics to represent and evolve behaviors and only discover behaviors in simulation, without testing or considering the deployment of these new behaviors on real robot swarms. In this work, we present Real2Sim2Real Behavior Discovery via Self-Supervised Representation Learning, which combines representation learning and novelty search to discover possible emergent behaviors automatically in simulation and enable direct controller transfer to real robots. First, we evaluate our method in simulation and show that our proposed self-supervised representation learning approach outperforms previous hand-crafted metrics by more accurately representing the space of possible emergent behaviors. Then, we address the reality gap by incorporating recent work in sim2real transfer for swarms into our lightweight simulator design, enabling direct robot deployment of all behaviors discovered in simulation on an open-source and low-cost robot platform.