Motion Planning Diffusion: Learning and Adapting Robot Motion Planning with Diffusion Models

TL;DR

This paper introduces Motion Planning Diffusion (MPD), a novel approach that leverages diffusion models to learn and adapt trajectory priors for robot motion planning, improving efficiency and smoothness in complex scenarios.

Contribution

The work presents a diffusion-based method for learning trajectory priors using low-dimensional B-spline representations, enabling more efficient and smooth robot motion planning.

Findings

Effective in 2D and complex 7-DOF tasks

Utilizes human demonstrations for real-world adaptation

Produces smooth, high-frequency interpolated trajectories

Abstract

The performance of optimization-based robot motion planning algorithms is highly dependent on the initial solutions, commonly obtained by running a sampling-based planner to obtain a collision-free path. However, these methods can be slow in high-dimensional and complex scenes and produce non-smooth solutions. Given previously solved path-planning problems, it is highly desirable to learn their distribution and use it as a prior for new similar problems. Several works propose utilizing this prior to bootstrap the motion planning problem, either by sampling initial solutions from it, or using its distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we introduce Motion Planning Diffusion (MPD), an algorithm that learns trajectory distribution priors with diffusion models. These generative models have shown increasing success in encoding multimodal data and have desirable properties for gradient-based motion planning, such as cost guidance. Given a motion planning problem, we construct a cost function and sample from the posterior distribution using the learned prior combined with the cost function gradients during the denoising process. Instead of learning the prior on all trajectory waypoints, we propose learning a lower-dimensional representation of a trajectory using linear motion primitives, particularly B-spline curves. This parametrization guarantees that the generated trajectory is smooth, can be interpolated at higher frequencies, and needs fewer parameters than a dense waypoint representation. We demonstrate the results of our method ranging from simple 2D to more complex tasks using a 7-dof robot arm manipulator. In addition to learning from simulated data, we also use human demonstrations on a real-world pick-and-place task.