[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-229037-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-229037-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","state-supervised-steering-function-for-sampling-based-kinodynamic-planning","State Supervised Steering Function for Sampling-based Kinodynamic Planning","","Sampling-based motion planners such as RRT* and BIT* depend on steering functions to connect sampled states with time-optimal motions in kinodynamic settings. Exact steering functions usually require either analytical solutions to the underlying time-optimal control problem or nonlinear programming solvers for the resulting boundary value problem, both of which limit real-time performance. This work introduces State Supervised Steering Function (S3F), learning near-optimal steering functions orders of magnitude faster than NLP baselines. Experiments on three robot domains show improved cost and runtime with RRT* using S3F, plus a probabilistic completeness result.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/state-supervised-steering-function-for-sampling-based-kinodynamic-planning/229037/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/state-supervised-steering-function-for-sampling-based-kinodynamic-planning/229037.png","ImageObject",300,407,{"name":42,"@type":43},"Elsa","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-22","2026-09-10",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",10,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why are steering functions difficult to use for time-optimal kinodynamic planning?","Question",{"text":63,"@type":64},"Exact steering functions typically require analytical solutions to the time-optimal control problem or nonlinear programming solvers for the boundary value problem. Both options are often unavailable for real domains or computationally too expensive.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What is S3F and how does it improve steering-function computation?",{"text":68,"@type":64},"S3F is a learned State Supervised Steering Function that produces near-optimal steering functions far faster than nonlinear programming counterparts. This enables efficient use of RRT* for kinodynamic planning.",{"name":70,"@type":61,"acceptedAnswer":71},"How is the method evaluated and what theoretical guarantee is provided?",{"text":72,"@type":64},"Experiments on three challenging robot domains compare cost and runtime, showing that RRT* with S3F outperforms state-of-the-art approaches. The paper also proves probabilistic completeness for RRT* modified to use S3F.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},229037,1789035129,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":55,"slug":125},"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":80,"read_time":144,"preview_page_count":4},137455077381,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","State Supervised Steering Function for Sampling-based  \nKinodynamic Planning  \nPranav Atreya  \nUniversity of Texas at Austin Austin, TX, United States [pranavatreya@utexas.edu](pranavatreya@utexas.edu)  \nJoydeep Biswas  \nUniversity of Texas at Austin Austin, TX, United States [joydeepb@cs.utexas.edu](joydeepb@cs.utexas.edu)  \nABSTRACT  \nSampling-based motion planners such as RRT* and BIT*, when applied to kinodynamic motion planning, rely on steering functions to generate time-optimal solutions connecting sampled states. Implementing exact steering functions requires either analytical solutions to the time-optimal control problem, or nonlinear programming (NLP) solvers to solve the boundary value problem given the system’s kinodynamic equations. Unfortunately, analytical solutions are unavailable for many real-world domains, and NLP solvers are prohibitively computationally expensive, hence fast and optimal kinodynamic motion planning remains an open problem. We provide a solution to this problem by introducing State Supervised Steering Function (S3F), a novel approach to learn time-optimal steering functions. S3F is able to produce near-optimal solutions to the steering function orders of magnitude faster than its NLP counterpart. Experiments conducted on three challenging robot domains show that RRT* using S3F significantly outperforms stateof-the-art planning approaches on both solution cost and runtime. We further provide a proof of probabilistic completeness of RRT* modified to use S3F.  \nKEYWORDS  \nKinodynamic Motion Planning; Learning Steering Functions; Sampling-based Planning  \nACM Reference Format:  \nPranavAtreya and Joydeep Biswas. 2022. State Supervised Steering Function for Sampling-based Kinodynamic Planning. In Proc. of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022), Online, May 9–13, 2022, IFAAMAS, 9 pages.  \n1 INTRODUCTION AND RELATED WORK  \nThis work tackles the kinodynamic motion planning (KDMP) problem, which is the problem of computing a kinodynamically feasible motion plan that takes a robot from an initial configuration to agoal region. We begin by formally defining the KDMP problem and then survey the various approaches to solving it.  \nLet 􀀭􀀘 be the configuration space of the robot. The state space 􀀭 is defined as the Cartesian product of 􀀭􀀘 with 􀀭􀀙, the set of dynamics variables needed to fully describe the dynamics of the robot at any given instance in time. 􀀭􀀙 typically consists of time derivatives of elements of 􀀭􀀘 . Let 􀀪 be the control space of the robot. The kinodynamic constraints are described by the differential equation 􀁇¤(􀁃) = 􀀵 (􀁇(􀁃),􀁄 (􀁃)), where 􀁇(􀁃) ∈ 􀀭 and 􀁄 (􀁃) ∈ 􀀪 . The  \nProc. of the 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022), P. Faliszewski, V. Mascardi, C. Pelachaud, M.E. Taylor (eds.), May 9–13, 2022, Online. © 2022 International Foundation for Autonomous Agents and Multiagent Systems ([www.ifaamas.org](www.ifaamas.org)). All rights reserved.  \nKDMP problem differs from the purely kinematic motion planning (KMP) problem in that the KMP problem operates only on the configuration space 􀀭􀀘. Let 􀀭obs ∈ 􀀭 be the set of obstacle-colliding states and let 􀀭free = 􀀭 \\􀀭obs be the set of valid states. Let 􀁇init ∈􀀭free be the initial state of the robot and let 􀀭goal ⊂ 􀀭free be the goal region. The objective of the KDMP problem is to find a collision free path that takes the robot from 􀁇init to 􀀭goal while obeying the kinodynamic constraints. The solution to the KDMP problem is a mapping 􀀲 (􀁃) : [0, 􀁃􀀵] → 􀀪 from time to control inputs such that applying 􀀲 (􀁃) starting from the state 􀁇init traces out a path 􀁢(􀁃) : [0, 􀁃􀀵] → 􀀭free such that 􀁢(􀁃􀀵) ∈ 􀀭goal. A motion plan is considered optimal if it minimizes some cost function 􀀘(􀁃􀀵, 􀀲, 􀁢) . The time-optimal solution minimizes the total time 􀁃􀀵 .  \nWe review the state ofthe art approaches to solving the KDMP problem, including search-based planning, sampling-ba","cbCaivNMZQEyhX0D","https://ap.wps.com/l/cbCaivNMZQEyhX0D","pdf",4928664,11,"English","# Abstract\n# Introduction and Related Work\n## Kinodynamic Motion Planning (KDMP) Formulation\n## State-of-the-Art Approaches","[{\"question\":\"Why are steering functions difficult to use for time-optimal kinodynamic planning?\",\"answer\":\"Exact steering functions typically require analytical solutions to the time-optimal control problem or nonlinear programming solvers for the boundary value problem. Both options are often unavailable for real domains or computationally too expensive.\"},{\"question\":\"What is S3F and how does it improve steering-function computation?\",\"answer\":\"S3F is a learned State Supervised Steering Function that produces near-optimal steering functions far faster than nonlinear programming counterparts. This enables efficient use of RRT* for kinodynamic planning.\"},{\"question\":\"How is the method evaluated and what theoretical guarantee is provided?\",\"answer\":\"Experiments on three challenging robot domains compare cost and runtime, showing that RRT* with S3F outperforms state-of-the-art approaches. The paper also proves probabilistic completeness for RRT* modified to use S3F.\"}]","State Supervised Steering Function for Sampling-based Kinodynamic Planning | PDF",28]