Diffusion models have recently been successfully applied to a wide range of robotics applications for learning complex multi-modal behaviors from data. However, prior works have mostly been confined to single-robot and small-scale environments due to the high sample complexity of learning multi-robot diffusion models. In this paper, we propose a method for generating collision-free multi-robot trajectories that conform to underlying data distributions while using only single-robot data. Our algorithm, Multi-robot Multi-model planning Diffusion (MMD), does so by combining learned diffusion models with classical search-based techniques -- generating data-driven motions under collision constraints. Scaling further, we show how to compose multiple diffusion models to plan in large environments where a single diffusion model fails to generalize well. We demonstrate the effectiveness of our approach in planning for dozens of robots in a variety of simulated scenarios motivated by logistics environments. View video demonstrations in our supplementary material, and our code at: github.com/yoraish/mmd.
One of the core ideas behind MMD is simple and effective: since diffusion models can admit soft-constraints via guidance functions within their denoising process, we can coordinate robots planning with diffusion models by using ideas borrowed from constraint-based multi-agent planning algorithms. In this paper we show that MMD is able to not only generate collision-free trajectories for dozens of robots, but, importantly, that the robots remain adherent to their individually-learned data distributions even when constrained. This allows for the generation of data-driven, collision-free trajectories for multiple robots in complex environments using only single-robot data.
@inproceedings{shaoul2024multi,
title={Multi-Robot Motion Planning with Diffusion Models},
author={Shaoul, Yorai and Mishani, Itamar and Vats, Shivam and Li, Jiaoyang and Likhachev, Maxim},
journal={The Thirteenth International Conference on Learning Representations (ICLR),
also at AAAI 2025 Workshop on Multi-Agent Path Finding},
year={2025},
}