[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85759-en":3,"doc-seo-85759-105":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},85759,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","Diffusion for Long-Horizon Multi-Robot Path Planning in Human-Shared Environments","Multi-robot path planning in human-shared environments demands both reliable inter-robot coordination and socially appropriate behavior around people. Existing diffusion-based generative planners often assume fixed-duration trajectories and incur high computation, limiting adaptation to changing goal distances and real-time use. Multi-Robot Rolling Diffusion (MRRD) enables real-time, long-horizon navigation for large robot teams in dense crowds by combining rolling-horizon planning, parallelized diffusion path generation, and conflict-based collision resolution. It adds urgency-based temporal conditioning and differentiated guidance to balance safety and efficient coordination, scaling to 15 robots in real time.","Diffusion for Long-Horizon Multi-Robot Path Planning in Human-Shared Environments  \nVaibhav Sanjay 1 , Yorai Shaoul 1 , and Jiaoyang Li 1  \narXiv :2607 .09911v1 [ cs .RO] 10 Jul 2026  \nAbstract—Multi-robot path planning in human-shared environments requires a delicate balance between robust inter-robot coordination and socially aware behavior. While diffusion models excel at generating predictable, human-like paths, existing generative planners are often restricted to paths of fixed duration and high computational latency, limiting their adaptability to varying goal distances and hindering real-time deployment. We present Multi-Robot Rolling Diffusion (MRRD), a novel framework that enables real-time, long-horizon navigation for large robot teams through dense crowds. MRRD combines a rolling-horizon scheme to accommodate the limited prediction horizon of human motion, parallelized diffusion inference for scalable generation of human-like paths, and a conflict-basedsearch mechanism for resolving inter-robot collisions. It further incorporates urgency-based temporal conditioning to generate paths with varying speeds and employs differentiated guidance terms to maximize both social awareness around humans and efficient coordination between robots. Experimental results in crowded environments demonstrate that MRRD successfully scales to 15 robots in real-time, significantly outperforming existing baselines in both safety and mission success rates.  \nI. INTRODUCTION  \nMulti-robot autonomous systems are successfully used in scenarios like warehouse automation, drone fleets, and inspection systems. However, their use in human-shared environments remains largely underexplored. This requires ensuring not only collision-free coordination among robots but also safe interaction with surrounding humans.  \nSingle-robot navigation in human-shared environments has been explored extensively [1]–[12], with a wide range of methods including human motion prediction, control barrier functions, and vision-language models. In this work, we focus on methods that directly generate human-like paths because they enable robots to move in a predictable and socially acceptable manner. Diffusion models [13], which excel at learning from demonstrations, are therefore a natural choice for this setting. Recently, CoBL-Diffusion [14] trains diffusion models on large sets of human walking data and demonstrates effective navigation for single robots in humanshared environments. However, these generative planners are often subject to a “fixed-time problem,” where they are restricted to produce paths of fixed duration regardless of the actual distance to the goal, leading to erratic velocities and reduced navigation efficiency.  \nResearch on multi-robot path planning in human-shared environments is not as extensive and can be largely summarized in a few distinct approaches. Prioritized planning  \n1Vaibhav Sanjay, Yorai Shaoul, and Jiaoyang Li are with the Robotics Institute, Carnegie Mellon University.{vsanjay,yshaoul,[jiaoyanl](jiaoyanl}@andrew.cmu.edu)[}](jiaoyanl}@andrew.cmu.edu)[@andrew.cmu.edu](jiaoyanl}@andrew.cmu.edu)  \nmethods [15], [16] anticipate human motions and plan robot paths in a predefined sequence, avoiding humans and higherpriority robots. Although quite fast, planning in sequence may result in uncooperative behavior since high-priority robots do not take the paths of low-priority robots into account. Hierarchical methods [17], [18] plan global paths ignoring collisions and resolve them during execution by maneuvering robots around each other. Since collisions are resolved by local planning, these methods may lack the ability to perform long-term planning while maintaining social awareness. RL-based methods [19] can directly learn robot coordination policies. However, they often rely on carefully engineered reward functions, which may not capture the full range of desirable human-like behaviors.  \nThis paper introduces Multi-Robot Rolling Diffusio","cbCaicDYhkJcCaLy","https://ap.wps.com/l/cbCaicDYhkJcCaLy","pdf",476014,1,"English","en",105,"# Introduction\n# Related Works","[{\"question\":\"What problem does MRRD address in human-shared multi-robot navigation?\",\"answer\":\"MRRD targets the need for collision-free robot coordination while also producing socially aware motion for humans. It also addresses fixed-duration and latency limitations in existing generative planners.\"},{\"question\":\"How does MRRD enable long-horizon planning despite human motion prediction limits?\",\"answer\":\"MRRD uses a rolling-horizon scheme so the system can generate path segments within the limited prediction horizon while still supporting long-horizon navigation.\"},{\"question\":\"How are robot-robot collisions handled in MRRD?\",\"answer\":\"Robot-robot interactions are resolved using Conflict-Based Search (CBS), while robot-human interactions are handled by the diffusion model generating human-like paths.\"}]",1784206058,20,{"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},"diffusion-for-long-horizon-multi-robot-path-planning-in-human-shared-environments","",{"@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/diffusion-for-long-horizon-multi-robot-path-planning-in-human-shared-environments/85759/",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},"What problem does MRRD address in human-shared multi-robot navigation?","Question",{"text":74,"@type":75},"MRRD targets the need for collision-free robot coordination while also producing socially aware motion for humans. It also addresses fixed-duration and latency limitations in existing generative planners.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does MRRD enable long-horizon planning despite human motion prediction limits?",{"text":79,"@type":75},"MRRD uses a rolling-horizon scheme so the system can generate path segments within the limited prediction horizon while still supporting long-horizon navigation.",{"name":81,"@type":72,"acceptedAnswer":82},"How are robot-robot collisions handled in MRRD?",{"text":83,"@type":75},"Robot-robot interactions are resolved using Conflict-Based Search (CBS), while robot-human interactions are handled by the diffusion model generating human-like paths.","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"]