Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models

TL;DR

This paper introduces Motion Planning Diffusion, a novel approach that employs diffusion models as priors for robot trajectory generation, enabling efficient and flexible motion planning especially in complex, high-dimensional environments.

Contribution

It proposes using diffusion models as priors for robot motion planning, allowing direct sampling from the posterior conditioned on task goals, and demonstrates superior performance over baselines.

Findings

Diffusion models effectively encode high-dimensional trajectory distributions.

The method generalizes well to environments with unseen obstacles.

Experiments show improved planning efficiency and accuracy.

Abstract

Learning priors on trajectory distributions can help accelerate robot motion planning optimization. Given previously successful plans, learning trajectory generative models as priors for a new planning problem is highly desirable. Prior works propose several ways on utilizing this prior to bootstrapping the motion planning problem. Either sampling the prior for initializations or using the prior distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we propose learning diffusion models as priors. We then can sample directly from the posterior trajectory distribution conditioned on task goals, by leveraging the inverse denoising process of diffusion models. Furthermore, diffusion has been recently shown to effectively encode data multimodality in high-dimensional settings, which is particularly well-suited for large trajectory dataset. To demonstrate our method efficacy, we compare our proposed method - Motion Planning Diffusion - against several baselines in simulated planar robot and 7-dof robot arm manipulator environments. To assess the generalization capabilities of our method, we test it in environments with previously unseen obstacles. Our experiments show that diffusion models are strong priors to encode high-dimensional trajectory distributions of robot motions.