{"ID":2877290,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.21001","arxiv_id":"2508.21001","title":"Train-Once Plan-Anywhere Kinodynamic Motion Planning via Diffusion Trees","abstract":"Kinodynamic motion planning is concerned with computing collision-free trajectories while abiding by the robot's dynamic constraints. This critical problem is often tackled using sampling-based planners (SBPs) that explore the robot's high-dimensional state space by constructing a search tree via action propagations. Although SBPs can offer global guarantees on completeness and solution quality, their performance is often hindered by slow exploration due to uninformed action sampling. Learning-based approaches can yield significantly faster runtimes, yet they fail to generalize to out-of-distribution (OOD) scenarios and lack critical guarantees, e.g., safety, thus limiting their deployment on physical robots. We present Diffusion Tree (DiTree): a provably-generalizable framework leveraging diffusion policies (DPs) as informed samplers to efficiently guide state-space search within SBPs. DiTree combines DP's ability to model complex distributions of expert trajectories, conditioned on local observations, with the completeness of SBPs to yield provably-safe solutions within a few action propagation iterations for complex dynamical systems. We demonstrate DiTree's power with an implementation combining the popular RRT planner with a DP action sampler trained on a single environment. In comprehensive evaluations on OOD scenarios, DiTree achieves on average a 30% higher success rate compared to standalone DP or SBPs, on a dynamic car and Mujoco's ant robot settings (for the latter, SBPs fail completely). Beyond simulation, real-world car experiments confirm DiTree's applicability, demonstrating superior trajectory quality and robustness even under severe sim-to-real gaps. Project webpage: https://sites.google.com/view/ditree.","short_abstract":"Kinodynamic motion planning is concerned with computing collision-free trajectories while abiding by the robot's dynamic constraints. This critical problem is often tackled using sampling-based planners (SBPs) that explore the robot's high-dimensional state space by constructing a search tree via action propagations. A...","url_abs":"https://arxiv.org/abs/2508.21001","url_pdf":"https://arxiv.org/pdf/2508.21001v2","authors":"[\"Yaniv Hassidof\",\"Tom Jurgenson\",\"Kiril Solovey\"]","published":"2025-08-28T17:04:00Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\",\"cs.RO\"]","methods":"[\"Diffusion Model\",\"LoRA\"]","project_urls":"[\"https://sites.google.com/view/ditree\"]","has_code":false,"code_links":[{"ID":610364,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2877290,"paper_url":"https://arxiv.org/abs/2508.21001","paper_title":"Train-Once Plan-Anywhere Kinodynamic Motion Planning via Diffusion Trees","repo_url":"https://github.com/google/safevalues","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}