[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82173-en":3,"doc-seo-82173-105":29,"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":21,"is_downloadable":21,"audit_status":21,"page_count":11,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},82173,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Dec-MARVEL Decentralized Multi-Agent Exploration without Communication under Budget Constraints","Dec-MARVEL is a decentralized budget-aware framework for multi-agent exploration when explicit communication is unreliable or unavailable. Instead of exchanging maps, goals, or messages, robots coordinate using incidental observations: any teammate trajectory appearing within an agent’s field of view becomes a coordination signal. A graph-attention actor fuses local frontier geometry, teammate motion, and budget features to select return-feasible waypoint-heading actions. Trained with phase-conditioned critics and a budget curriculum, it achieves top or tied-best exploration rates with reduced sensing overlap across 900 held-out trials and demonstrates successful sim-to-real transfer.","Dec-MARVEL: Decentralized Multi-Agent Exploration without Communication under Budget Constraints  \nJanghyun Cho 1 ,∗ , Jimmy Chiun2 ,∗ , Guillaume Sartoretti2 , and Changjoo Nam3 ,†  \narXiv :2607 .09060v1 [ cs .RO] 10 Jul 2026  \nAbstract—Multi-UAV exploration is often constrained by unreliable communication, limited field-of-view sensing (e.g., lightweight onboard camera), and finite travel budgets that require each robot to reserve enough budget to return to its base. We present Dec-MARVEL, a decentralized budgetaware exploration framework for communication-free teams with directional sensing. Rather than exchanging maps, goals, or messages, each robot coordinates through its incidental observations: any teammate trajectory within its field of view serves as a coordination signal. A graph-attention actor fuses local frontier geometry, teammate motion, and budget features to select return-feasible waypoint-heading actions. The actor is trained with phase-conditioned critics, a training-only taskoriented privileged critic, and a mixture-based budget curriculum. Across 900 held-out trials spanning three team sizes (2, 4, 8 robots) and three travel budgets (720, 800, 1024 meters) against four baselines, Dec-MARVEL achieves the highest or tiedhighest exploration rate and lowest sensing overlap across all nine team-size × budget configurations. Under our tightest 720m budget, it reaches 53%, 94%, and 100% success for 2, 4, and 8 robots, versus 37%, 83%, and 99% for the strongest baseline. Physical-robot experiments demonstrate successful sim-to-real transfer and real-world deployment of Dec-MARVEL.  \nI. INTRODUCTION  \nAutonomous exploration of unknown environments is a fundamental problem in field robotics, with applications in search and rescue, infrastructure inspection, and hazardousenvironment surveys. Using multiple robots can improve exploration efficiency by covering different regions in parallel [1] . However, this benefit depends strongly on coordination: if multiple robots explore independently, the team would have high overlap and gain limited performance improvements over parallel individual exploration. Many existing multi-robot exploration methods address this issue by relying on explicit communication, such as sharing maps, exchanging goals, or assigning frontiers among robots [2],[3] . In practical field settings, especially on uncrewed aerial vehicles (UAVs), this assumption can be difficult to satisfy because wireless communication can be degraded or unavailable due to occlusions, signal attenuation, limited bandwidth, or infrastructure damage [4], [5] .  \nIn this work, we consider a different form of coordination. Even when robots do not exchange messages, they may still occasionally observe one another through their onboard  \n†Equal contribution.  \n∗ Corresponding author: [cjnam@sogang.ac.kr](cjnam@sogang.ac.kr)  \n[1](1 Dept.)[ Dept.](1 Dept.) of Artificial Intelligence, Sogang University.  \n2Dept. of Mechanical Engineering, College of Design and Engineering, National University of Singapore.  \n3Dept. of Electronic Engineering, Sogang University.  \n Agent  Observed Trajectory  Sensing Area(FoV)  \nFig. 1: Overview of Dec-MARVEL for distributed budget-aware multi-agent exploration without explicit inter-robot communication. Dec-MARVEL enables decentralized communication-free multi-robot exploration through opportunistic trajectory observation. By observing teammate trajectory within their FoV, agents infer recently explored directions and coordinate waypointheading decisions without explicit communication.  \nsensors. Prior work on vision-based and limited Field-ofView (FoV) coordination suggests that such visibility constraints can be used as part of the coordination process [6]–[8] . These opportunistic observations (Fig. 1) are limited and intermittent, but we show that they can still provide useful information about where a teammate has recently moved, and thus implicitly about which regions may have already be","cbCainzig8l2mbmJ","https://ap.wps.com/l/cbCainzig8l2mbmJ","pdf",2843643,3,1,"English","en",105,"# Introduction\n## Problem setting: multi-robot exploration and overlap\n## Challenge: communication constraints in UAVs\n## Coordination via opportunistic teammate trajectory observation","[{\"question\":\"What coordination signal does Dec-MARVEL use when robots cannot communicate?\",\"answer\":\"Dec-MARVEL uses opportunistically observed teammate trajectories within an agent’s field of view. Each observed trajectory indicates recently explored directions, enabling implicit coordination without map or message exchange.\"},{\"question\":\"How does Dec-MARVEL ensure robots can still return to base under limited travel budgets?\",\"answer\":\"The method models remaining travel budget and selects waypoint-heading actions that are feasible for returning to base. A budget-aware mechanism prevents overly aggressive motion that would risk exhausting budget.\"},{\"question\":\"What performance results does Dec-MARVEL report compared with baselines?\",\"answer\":\"Across 900 held-out trials for team sizes 2, 4, and 8 and travel budgets 720, 800, and 1024 meters, Dec-MARVEL achieves the highest or tied-highest exploration rate and the lowest sensing overlap across all nine configurations. Under the tightest 720m budget, it reaches 53%, 94%, and 100% success for 2, 4, and 8 robots.\"}]",1784178591,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"dec-marvel-decentralized-multi-agent-exploration-without-communication-under-budget-constraints","",{"@graph":35,"@context":84},[36,52,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":20},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/dec-marvel-decentralized-multi-agent-exploration-without-communication-under-budget-constraints/82173/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-22","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 coordination signal does Dec-MARVEL use when robots cannot communicate?","Question",{"text":74,"@type":75},"Dec-MARVEL uses opportunistically observed teammate trajectories within an agent’s field of view. Each observed trajectory indicates recently explored directions, enabling implicit coordination without map or message exchange.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does Dec-MARVEL ensure robots can still return to base under limited travel budgets?",{"text":79,"@type":75},"The method models remaining travel budget and selects waypoint-heading actions that are feasible for returning to base. A budget-aware mechanism prevents overly aggressive motion that would risk exhausting budget.",{"name":81,"@type":72,"acceptedAnswer":82},"What performance results does Dec-MARVEL report compared with baselines?",{"text":83,"@type":75},"Across 900 held-out trials for team sizes 2, 4, and 8 and travel budgets 720, 800, and 1024 meters, Dec-MARVEL achieves the highest or tied-highest exploration rate and the lowest sensing overlap across all nine configurations. Under the tightest 720m budget, it reaches 53%, 94%, and 100% success for 2, 4, and 8 robots.","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":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":21,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]