Large Scale Distributed Collaborative Unlabeled Motion Planning with Graph Policy Gradients

TL;DR

This paper introduces a scalable multi-robot motion planning method using graph neural networks that efficiently handles large robot teams with arbitrary dynamics and constraints, enabling zero-shot policy transfer.

Contribution

The paper proposes a novel graph neural network-based framework for large-scale multi-robot motion planning that allows for scalable training and zero-shot policy transfer to larger robot groups.

Findings

Effective in simulations with large robot teams

Enables zero-shot transfer of policies to more robots

Handles arbitrary dynamics and constraints

Abstract

In this paper, we present a learning method to solve the unlabelled motion problem with motion constraints and space constraints in 2D space for a large number of robots. To solve the problem of arbitrary dynamics and constraints we propose formulating the problem as a multi-agent problem. We are able to demonstrate the scalability of our methods for a large number of robots by employing a graph neural network (GNN) to parameterize policies for the robots. The GNN reduces the dimensionality of the problem by learning filters that aggregate information among robots locally, similar to how a convolutional neural network is able to learn local features in an image. Additionally, by employing a GNN we are also able to overcome the computational overhead of training policies for a large number of robots by first training graph filters for a small number of robots followed by zero-shot policy transfer to a larger number of robots. We demonstrate the effectiveness of our framework through various simulations.