{"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/hyp-despot-a-hybrid-parallel-algorithm-for","title":"HyP-DESPOT: A Hybrid Parallel Algorithm for Online Planning under Uncertainty","arxiv_id":"1802.06215","date":"2018-02-17","proceeding":null,"authors":["Panpan Cai","Yuanfu Luo","David Hsu","Wee Sun Lee"],"abstract":"Planning under uncertainty is critical for robust robot performance in\nuncertain, dynamic environments, but it incurs high computational cost.\nState-of-the-art online search algorithms, such as DESPOT, have vastly improved\nthe computational efficiency of planning under uncertainty and made it a\nvaluable tool for robotics in practice. This work takes one step further by\nleveraging both CPU and GPU parallelization in order to achieve near real-time\nonline planning performance for complex tasks with large state, action, and\nobservation spaces. Specifically, we propose Hybrid Parallel DESPOT\n(HyP-DESPOT), a massively parallel online planning algorithm that integrates\nCPU and GPU parallelism in a multi-level scheme. It performs parallel DESPOT\ntree search by simultaneously traversing multiple independent paths using\nmulti-core CPUs and performs parallel Monte-Carlo simulations at the leaf nodes\nof the search tree using GPUs. Experimental results show that HyP-DESPOT speeds\nup online planning by up to several hundred times, compared with the original\nDESPOT algorithm, in several challenging robotic tasks in simulation.","url_abs":"http://arxiv.org/abs/1802.06215v1","url_pdf":"http://arxiv.org/pdf/1802.06215v1.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":"hyp-despot-a-hybrid-parallel-algorithm-for","repo_url":"https://github.com/AdaCompNUS/hyp-despot","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":null,"task_name":"GPU"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.06215","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}