[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82495-en":3,"doc-seo-82495-105":28,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":13,"seo_description":14,"update_tm":26,"read_time":27},82495,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","NeHMO Neural Hamilton-Jacobi Reachability Learning for Decentralized Safe Multi-Arm Motion Planning","Safe multi-arm motion planning is difficult in robotics because high-dimensional coupled configurations and complex collision constraints make real-time coordination hard. Centralized planners can coordinate all arms but often do not scale, while decentralized methods may rely on behavior prediction or coordination assumptions that fail under uncertainty. The NeHMO approach introduces neural Hamilton-Jacobi Reachability learning to approximate a worst-case safety value function, then uses it in decentralized trajectory optimization for real-time planning. The method is scalable, data-efficient, generalizes across multi-manipulator systems, and outperforms baselines on challenging tasks.","NeHMO: Neural Hamilton-Jacobi Reachability Learning for Decentralized Safe Multi-Arm Motion Planning  \nQingyi Chen 1 , Zachary Kingston 1 and Ahmed H. Qureshi 1  \narXiv :2607 .00326v 1 [ cs .RO] 1 Jul 2026  \nAbstract—Safe multi-arm motion planning is a challenging problem in robotics due to its high dimensionality, coupled configuration space, and complex collision constraints. Centralized planners are capable of coordinating all arms but often face scalability limitations, restricting applicability in realtime settings. On the other hand, decentralized methods are scalable and recent deep learning-based approaches have shown promising results. However, these depend on accurate behavior prediction or coordination protocols and may fail when other arms act unpredictably. To address these challenges, we introduce a neural Hamilton-Jacobi Reachability (HJR) learningbased approach to approximate a safety value function that captures worst-case inter-arm safety constraints. We further develop a decentralized trajectory optimization framework that uses the learned HJR representation for real-time planning. The proposed method is scalable and data-efficient, generalizes across multi-manipulator systems, and outperforms state-ofthe-art baselines on challenging multi-arm motion planning tasks.  \nI. INTRODUCTION  \nMulti-arm motion planning plays an important role in robotic manipulation systems, including coordinated assembly, shared-workspace manipulation, and warehouse automation. However, ensuring system-wide safety while generating motion plans in real time remains a significant challenge.  \nOne approach is to plan jointly for all robots using a centralized controller [1]–[3] . Such planners explicitly coordinate all arms and enforce global safety constraints. However, their computational complexity grows rapidly with the number of manipulators [4], limiting scalability and realtime applicability. An alternative is decentralized planning, where each arm generates its own plan, improving scalability and computational efficiency at the cost of typically relying on predicting other arms’ behaviors or assuming their compliance with a shared control policy. These assumptions may break down in unstructured or safety-critical settings, particularly when other manipulators behave unpredictably. Hamilton-Jacobi Reachability (HJR) provides a principled framework for modeling safety-critical control and has been increasingly applied to multi-robot systems [5]–[7] . Based on the Hamilton-Jacobi-Isaacs (HJI) equation, HJR formulates safety as a differential game between control and disturbance, enabling reasoning about worst-case interactions under dynamic constraints. In multi-arm systems, this formulation naturally captures inter-arm collision avoidance under uncertainty or adversarial behavior. However, classical HJR  \n1 Qingyi Chen, Zachary Kingston and Ahmed H. Qureshi are with the Department of Computer Science at Purdue University, West Lafayette, IN, USA. {chen5221, zkingston, [ahqureshi](ahqureshi}@purdue.edu)[}](ahqureshi}@purdue.edu)[@purdue.edu](ahqureshi}@purdue.edu).  \n(1) (2)  \n(3) (4)  \nFig. 1: Our method, NeHMO, controlling a 12-dimensional dual-UR5 system. The images show the two manipulators getting close to each other and adjusting their configurations to avoid a potential collision.  \nmethods suffer from exponential scaling with dimensionality, making direct application to high-dimensional systems infeasible. To address this limitation, Bansal and Tomlin [8] introduced DeepReach, which uses neural networks to approximate HJR solutions through self-supervised training. While this approach removes the need for labeled data, it requires extensive training and does not readily scale to highdimensional multi-arm systems.  \nTo address these challenges, we present Neural HJRguided Multi-arm Motion Optimizer (NeHMO) . NeHMO augments DeepReach with physics priors in multi-robot systems, improving the modeling and scalability of neura","cbCaikjlu5210zdm","https://ap.wps.com/l/cbCaikjlu5210zdm","pdf",10378211,1,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"Why do centralized planners struggle with safe multi-arm motion planning in real time?\",\"answer\":\"Centralized planners explicitly coordinate all arms and enforce global safety, but their computational complexity grows rapidly with the number of manipulators, limiting scalability for real-time use.\"},{\"question\":\"What problem do decentralized methods face when other arms act unpredictably?\",\"answer\":\"Decentralized approaches often depend on predicting other arms’ behaviors or assuming coordination compliance. These assumptions can break down in unstructured, safety-critical settings with adversarial or unpredictable actions.\"},{\"question\":\"How does NeHMO improve real-time decentralized safety planning?\",\"answer\":\"NeHMO learns a Hamilton-Jacobi Reachability-based worst-case inter-arm safety value function and integrates the learned representation into a decentralized trajectory optimization framework to actively avoid unsafe interactions.\"}]",1784180927,20,{"code":4,"msg":29,"data":30},"ok",{"site_id":23,"language":22,"slug":31,"title":13,"keywords":32,"description":14,"schema_data":33,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":26},"nehmo-neural-hamilton-jacobi-reachability-learning-for-decentralized-safe-multi-arm-motion-planning","",{"@graph":34,"@context":84},[35,52,67],{"@type":36,"itemListElement":37},"BreadcrumbList",[38,42,46,49],{"item":39,"name":40,"@type":41,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":43,"name":44,"@type":41,"position":45},"https://docshare.wps.com/document/","Document",2,{"item":47,"name":12,"@type":41,"position":48},"https://docshare.wps.com/document/research-report/",3,{"item":50,"name":13,"@type":41,"position":51},"https://docshare.wps.com/document/nehmo-neural-hamilton-jacobi-reachability-learning-for-decentralized-safe-multi-arm-motion-planning/82495/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":39,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why do centralized planners struggle with safe multi-arm motion planning in real time?","Question",{"text":74,"@type":75},"Centralized planners explicitly coordinate all arms and enforce global safety, but their computational complexity grows rapidly with the number of manipulators, limiting scalability for real-time use.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What problem do decentralized methods face when other arms act unpredictably?",{"text":79,"@type":75},"Decentralized approaches often depend on predicting other arms’ behaviors or assuming coordination compliance. These assumptions can break down in unstructured, safety-critical settings with adversarial or unpredictable actions.",{"name":81,"@type":72,"acceptedAnswer":82},"How does NeHMO improve real-time decentralized safety planning?",{"text":83,"@type":75},"NeHMO learns a Hamilton-Jacobi Reachability-based worst-case inter-arm safety value function and integrates the learned representation into a decentralized trajectory optimization framework to actively avoid unsafe interactions.","https://schema.org",{"og:url":50,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":44,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":45,"doc_module":4,"doc_module_name":44,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":44,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":44,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":44,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":44,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":44,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":44,"category_name":124,"show_sort_weight":27,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":27,"doc_module":4,"doc_module_name":44,"category_name":127,"show_sort_weight":27,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":44,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":44,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]