{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-sampling-distributions-for-robot","title":"Learning Sampling Distributions for Robot Motion Planning","arxiv_id":"1709.05448","date":"2017-09-16","proceeding":null,"authors":["Brian Ichter","James Harrison","Marco Pavone"],"abstract":"A defining feature of sampling-based motion planning is the reliance on an\nimplicit representation of the state space, which is enabled by a set of\nprobing samples. Traditionally, these samples are drawn either\nprobabilistically or deterministically to uniformly cover the state space. Yet,\nthe motion of many robotic systems is often restricted to \"small\" regions of\nthe state space, due to, for example, differential constraints or\ncollision-avoidance constraints. To accelerate the planning process, it is thus\ndesirable to devise non-uniform sampling strategies that favor sampling in\nthose regions where an optimal solution might lie. This paper proposes a\nmethodology for non-uniform sampling, whereby a sampling distribution is\nlearned from demonstrations, and then used to bias sampling. The sampling\ndistribution is computed through a conditional variational autoencoder,\nallowing sample generation from the latent space conditioned on the specific\nplanning problem. This methodology is general, can be used in combination with\nany sampling-based planner, and can effectively exploit the underlying\nstructure of a planning problem while maintaining the theoretical guarantees of\nsampling-based approaches. Specifically, on several planning problems, the\nproposed methodology is shown to effectively learn representations for the\nrelevant regions of the state space, resulting in an order of magnitude\nimprovement in terms of success rate and convergence to the optimal cost.","url_abs":"http://arxiv.org/abs/1709.05448v3","url_pdf":"http://arxiv.org/pdf/1709.05448v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-sampling-distributions-for-robot","repo_url":"https://github.com/StanfordASL/LearnedSamplingDistributions","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-sampling-distributions-for-robot","repo_url":"https://github.com/RogerQi/CVAE_Motion_Planning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"collision-avoidance","task_name":"Collision Avoidance"},{"task_slug":"motion-planning","task_name":"Motion Planning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1709.05448","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}