DiffusionSeeder: Seeding Motion Optimization with Diffusion for Rapid Motion Planning

Huang Huang, Balakumar Sundaralingam, Arsalan Mousavian, Adithyavairavan Murali, Ken Goldberg, Dieter Fox
Proceedings of The 8th Conference on Robot Learning, PMLR 270:4392-4409, 2025.

Abstract

Running optimization across many parallel seeds leveraging GPU compute [2] have relaxed the need for a good initialization, but this can fail if the problem is highly non-convex as all seeds could get stuck in local minima. One such setting is collision-free motion optimization for robot manipulation, where optimization converges quickly on easy problems but struggle in obstacle dense environments (e.g., a cluttered cabinet or table). In these situations, graph based planning algorithms are called to obtain seeds, resulting significant slowdowns. We propose DiffusionSeeder, a diffusion based approach that generates trajectories to seed motion optimization for rapid robot motion planning. DiffusionSeeder takes the initial depth image observation of the scene and generates high quality, multi-modal trajectories that are then fine-tuned with few iterations of motion optimization. We integrated DiffusionSeeder with cuRobo, a GPU-accelerated motion optimization method, to generate the seed trajectories which results in 12x speed up on average, and 36x speed up for more complicated problems, while achieving 10% higher success rate in partially observed simulation environments. Our results prove the effectiveness of using diverse solutions from learned diffusion model. Physical experiments on a Franka robot demonstrate the sim2real transfer of DiffusionSeeder to the real robot, with an average success rate of 86% and planning time of 26ms, increasing on cuRobo by 51% higher success rate and 2.5x speed up. The code and the model weights will be available after publication.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-huang25f, title = {DiffusionSeeder: Seeding Motion Optimization with Diffusion for Rapid Motion Planning}, author = {Huang, Huang and Sundaralingam, Balakumar and Mousavian, Arsalan and Murali, Adithyavairavan and Goldberg, Ken and Fox, Dieter}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {4392--4409}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/huang25f/huang25f.pdf}, url = {https://proceedings.mlr.press/v270/huang25f.html}, abstract = {Running optimization across many parallel seeds leveraging GPU compute [2] have relaxed the need for a good initialization, but this can fail if the problem is highly non-convex as all seeds could get stuck in local minima. One such setting is collision-free motion optimization for robot manipulation, where optimization converges quickly on easy problems but struggle in obstacle dense environments (e.g., a cluttered cabinet or table). In these situations, graph based planning algorithms are called to obtain seeds, resulting significant slowdowns. We propose DiffusionSeeder, a diffusion based approach that generates trajectories to seed motion optimization for rapid robot motion planning. DiffusionSeeder takes the initial depth image observation of the scene and generates high quality, multi-modal trajectories that are then fine-tuned with few iterations of motion optimization. We integrated DiffusionSeeder with cuRobo, a GPU-accelerated motion optimization method, to generate the seed trajectories which results in 12x speed up on average, and 36x speed up for more complicated problems, while achieving 10% higher success rate in partially observed simulation environments. Our results prove the effectiveness of using diverse solutions from learned diffusion model. Physical experiments on a Franka robot demonstrate the sim2real transfer of DiffusionSeeder to the real robot, with an average success rate of 86% and planning time of 26ms, increasing on cuRobo by 51% higher success rate and 2.5x speed up. The code and the model weights will be available after publication.} }
Endnote
%0 Conference Paper %T DiffusionSeeder: Seeding Motion Optimization with Diffusion for Rapid Motion Planning %A Huang Huang %A Balakumar Sundaralingam %A Arsalan Mousavian %A Adithyavairavan Murali %A Ken Goldberg %A Dieter Fox %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-huang25f %I PMLR %P 4392--4409 %U https://proceedings.mlr.press/v270/huang25f.html %V 270 %X Running optimization across many parallel seeds leveraging GPU compute [2] have relaxed the need for a good initialization, but this can fail if the problem is highly non-convex as all seeds could get stuck in local minima. One such setting is collision-free motion optimization for robot manipulation, where optimization converges quickly on easy problems but struggle in obstacle dense environments (e.g., a cluttered cabinet or table). In these situations, graph based planning algorithms are called to obtain seeds, resulting significant slowdowns. We propose DiffusionSeeder, a diffusion based approach that generates trajectories to seed motion optimization for rapid robot motion planning. DiffusionSeeder takes the initial depth image observation of the scene and generates high quality, multi-modal trajectories that are then fine-tuned with few iterations of motion optimization. We integrated DiffusionSeeder with cuRobo, a GPU-accelerated motion optimization method, to generate the seed trajectories which results in 12x speed up on average, and 36x speed up for more complicated problems, while achieving 10% higher success rate in partially observed simulation environments. Our results prove the effectiveness of using diverse solutions from learned diffusion model. Physical experiments on a Franka robot demonstrate the sim2real transfer of DiffusionSeeder to the real robot, with an average success rate of 86% and planning time of 26ms, increasing on cuRobo by 51% higher success rate and 2.5x speed up. The code and the model weights will be available after publication.
APA
Huang, H., Sundaralingam, B., Mousavian, A., Murali, A., Goldberg, K. & Fox, D.. (2025). DiffusionSeeder: Seeding Motion Optimization with Diffusion for Rapid Motion Planning. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:4392-4409 Available from https://proceedings.mlr.press/v270/huang25f.html.

Related Material