DANCeRS: A Distributed Algorithm for Negotiating Consensus in Robot Swarms with Gaussian Belief Propagation
Aalok Patwardhan, Andrew J Davison
AI summary
Problem
Existing approaches treat consensus in discrete and continuous decision spaces as separate problems, limiting scalability and robustness in dynamic swarm coordination.
Approach
The method models the swarm as a factor graph and applies Gaussian Belief Propagation over Lie groups to enable purely distributed, peer-to-peer message passing for joint path planning and consensus negotiation.
Key results
- Unified framework for discrete and continuous consensus
- GBP extension for non-holonomic unicycle path planning
- Novel distributed shape formation with discrete target assignment
- First application of GBP for discrete consensus in robot swarms
Why it matters
Enables scalable, robust multi-robot coordination without centralized control, advancing practical applications in swarm robotics and distributed AI.
Abstract
Robot swarms require cohesive collective be- haviour to address diverse challenges, including shape for- mation and decision-making. Existing approaches often treat consensus in discrete and continuous decision spaces as dis- tinct problems. We present DANCeRS, a unified, distributed algorithm leveraging Gaussian Belief Propagation (GBP) to achieve consensus in both domains. By representing a swarm as a factor graph our method ensures scalability and robustness in dynamic environments, relying on purely peer-to-peer mes- sage passing. We demonstrate the effectiveness of our general framework through two applications where agents in a swarm must achieve consensus on global behaviour whilst relying on local communication. In the first, robots must perform path planning and collision avoidance to create shape formations. In the second, we show how the same framework can be used by a group of robots to form a consensus over a set of discrete decisions. Experimental results highlight our method’s scalability and efficiency compared to recent approaches to these problems making it a promising solution for multi-robot systems requiring distributed consensus. We encourage the reader to see the supplementary video demo.