LEMURS: Learning Distributed Multi-Robot Interactions

TL;DR

LEMURS introduces a scalable, distributed learning algorithm for multi-robot control that leverages physical constraints and advanced neural architectures to enable cooperative behaviors from demonstrations.

Contribution

The paper proposes a novel port-Hamiltonian based neural architecture combining self-attention and neural ODEs for scalable multi-robot control learning.

Findings

Successfully learned multi-agent navigation behaviors

Achieved stable cooperative flocking control

Demonstrated scalability across different robot team sizes

Abstract

This paper presents LEMURS, an algorithm for learning scalable multi-robot control policies from cooperative task demonstrations. We propose a port-Hamiltonian description of the multi-robot system to exploit universal physical constraints in interconnected systems and achieve closed-loop stability. We represent a multi-robot control policy using an architecture that combines self-attention mechanisms and neural ordinary differential equations. The former handles time-varying communication in the robot team, while the latter respects the continuous-time robot dynamics. Our representation is distributed by construction, enabling the learned control policies to be deployed in robot teams of different sizes. We demonstrate that LEMURS can learn interactions and cooperative behaviors from demonstrations of multi-agent navigation and flocking tasks.