SAGE: Decoupling Spatial Logic from Metric Scale for Zero-Shot Multi-Robot Exploration

preprint OA: closed CC-BY-4.0
AI-generated deep summary by claude@2026-07, 2026-07-03 · read from full text

This paper presents SAGE, a zero-shot multi-robot exploration framework for unknown, unbounded, irregular environments, where high-level coordination typically depends on global spatial priors or fixed metric-scale representations. The authors abstract high-dimensional sensor streams into a sparse, dynamic graph using topological invariants to decouple decision logic from physical scale, and they stabilize training with a capacity-controlled multi-agent reinforcement learning approach that uses dynamic masking to synchronize observable horizon with effective action space. Policies trained only on structured 60×60 grid worlds reportedly transfer zero-shot to environments with very different physics and scales, including large-scale unstructured maps and real-world dense, irregular forest settings. A major caveat explicitly implied by the preprint nature is that results have not been peer reviewed. The paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

Abstract Effective multi-robot exploration in unknown environments faces a critical structural dilemma: high-level coordination typically relies on global spatial priors or fixed-scale representations, yet real-world missions operate in unbounded, irregular domains where such priors are unavailable. Here, we present SAGE, a framework that resolves this dilemma by treating exploration as a topological invariant, independent of metric boundaries. SAGE acts as a dimensional reduction operator, abstracting high-dimensional sensor streams into a sparse, dynamic graph that decouples decision logic from physical scale. To stabilize learning on these continuously growing structures, we introduce a capacity-controlled multi-agent reinforcement learning regime. By synchronizing the agent's observable horizon with its effective action space via a dynamic masking protocol, this mechanism constrains the complexity of the growing environment, ensuring the policy masters local topological correlations. We demonstrate that policies trained exclusively on minimalist, structured 60 × 60 grid worlds transfer zero-shot to environments of vastly different physics and scales—from large-scale unstructured maps to real-world deployment in dense, irregular forests. These results verify that abstracting spatial logic from metric geometry enables robust, boundless exploration without environment-specific tuning.
Full text 13,635 characters · extracted from preprint-html · click to expand
SAGE: Decoupling Spatial Logic from Metric Scale for Zero-Shot Multi-Robot Exploration | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article SAGE: Decoupling Spatial Logic from Metric Scale for Zero-Shot Multi-Robot Exploration Peng Lu, Xiao Cao, Mingyang Li, Yi Luo, Yuting Tao, Bowen Xu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8307666/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Effective multi-robot exploration in unknown environments faces a critical structural dilemma: high-level coordination typically relies on global spatial priors or fixed-scale representations, yet real-world missions operate in unbounded, irregular domains where such priors are unavailable. Here, we present SAGE, a framework that resolves this dilemma by treating exploration as a topological invariant, independent of metric boundaries. SAGE acts as a dimensional reduction operator, abstracting high-dimensional sensor streams into a sparse, dynamic graph that decouples decision logic from physical scale. To stabilize learning on these continuously growing structures, we introduce a capacity-controlled multi-agent reinforcement learning regime. By synchronizing the agent's observable horizon with its effective action space via a dynamic masking protocol, this mechanism constrains the complexity of the growing environment, ensuring the policy masters local topological correlations. We demonstrate that policies trained exclusively on minimalist, structured 60 × 60 grid worlds transfer zero-shot to environments of vastly different physics and scales—from large-scale unstructured maps to real-world deployment in dense, irregular forests. These results verify that abstracting spatial logic from metric geometry enables robust, boundless exploration without environment-specific tuning. Physical sciences/Engineering/Mechanical engineering Physical sciences/Engineering/Aerospace engineering Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryInformation.pdf Supplementary Information finalmark1nc.mp4 Supplementary Movie 1 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8307666","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":560599400,"identity":"a0cf205a-e87d-4664-95fd-33170b785fd5","order_by":0,"name":"Peng Lu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAp0lEQVRIiWNgGAWjYBACPghlkyDBwMDG2ECMFjYIlUa6lsOkaJHIPfi44Nf5PMkZCWwPZxCnJS/ZeGbf7WJpiQR2ww3Eackxk+btuZ04TyKBTfIBkVrMf/P2nCNNixkzz48DibNBWohzGM+7ZGnehuTEmT0P2ySJ8j4/e+7Bzzx/7BJnHE8+JtlDjBYGBh4GBsY2EIO4WIFqYfhDrOJRMApGwSgYkQAAwMMvUA6ZztcAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-5611-4902","institution":"The University of Hong Kong","correspondingAuthor":true,"prefix":"","firstName":"Peng","middleName":"","lastName":"Lu","suffix":""},{"id":560599401,"identity":"ea4e5194-1802-44c3-846f-625b277f48ac","order_by":1,"name":"Xiao Cao","email":"","orcid":"","institution":"The University of Hong Kong (HKU)","correspondingAuthor":false,"prefix":"","firstName":"Xiao","middleName":"","lastName":"Cao","suffix":""},{"id":560599402,"identity":"d9ff3770-1904-4dfc-9590-7794e74f012d","order_by":2,"name":"Mingyang Li","email":"","orcid":"https://orcid.org/0009-0006-4536-4811","institution":"The University of Hong Kong (HKU)","correspondingAuthor":false,"prefix":"","firstName":"Mingyang","middleName":"","lastName":"Li","suffix":""},{"id":560599403,"identity":"7b04a182-c811-4675-a68b-dd79ac98170a","order_by":3,"name":"Yi Luo","email":"","orcid":"https://orcid.org/0009-0007-6817-9173","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Luo","suffix":""},{"id":560599404,"identity":"3e2b4e79-9dd2-4c22-a2f7-9bf5dd10f9aa","order_by":4,"name":"Yuting Tao","email":"","orcid":"","institution":"The University of Hong Kong (HKU)","correspondingAuthor":false,"prefix":"","firstName":"Yuting","middleName":"","lastName":"Tao","suffix":""},{"id":560599405,"identity":"e81d1a41-a15d-4007-85ee-ad7042537d3e","order_by":5,"name":"Bowen Xu","email":"","orcid":"","institution":"The University of Hong Kong (HKU)","correspondingAuthor":false,"prefix":"","firstName":"Bowen","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2025-12-08 12:24:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8307666/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8307666/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98291127,"identity":"00496cc5-40cd-4f9e-a3cf-19addff7d4fe","added_by":"auto","created_at":"2025-12-16 08:21:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39925800,"visible":true,"origin":"","legend":"","description":"","filename":"SAGE.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8307666/v1/d9d7a57e7aee9808ac25f76a.pdf"},{"id":98436078,"identity":"9e7e8926-265f-4705-92b0-df1f4d647782","added_by":"auto","created_at":"2025-12-17 16:54:52","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7336,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25099127.json","url":"https://assets-eu.researchsquare.com/files/rs-8307666/v1/21add064639522602b7916b2.json"},{"id":98291125,"identity":"5ec886a5-cb37-4582-ad8e-ba3fe8cf758a","added_by":"auto","created_at":"2025-12-16 08:21:24","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":46660,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8307666/v1/5796e3e8b21d1454b258842b.pdf"},{"id":98445524,"identity":"be4b2eeb-111f-48f8-ba23-576562657f6a","added_by":"auto","created_at":"2025-12-17 17:19:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8516920,"visible":true,"origin":"","legend":"Article File","description":"","filename":"SAGE.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8307666/v1_covered_940f7909-e2ba-48a8-8cc5-309fddb6a101.pdf"},{"id":98291124,"identity":"b3efab33-46ad-4f8a-8f63-f7e94b1167ea","added_by":"auto","created_at":"2025-12-16 08:21:24","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":46660,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8307666/v1/e87551837fb6a451fe2f2b71.pdf"},{"id":98291128,"identity":"378d5975-e223-424a-a73f-3f6fe6d3593b","added_by":"auto","created_at":"2025-12-16 08:21:25","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":95502441,"visible":true,"origin":"","legend":"Supplementary Movie 1","description":"","filename":"finalmark1nc.mp4","url":"https://assets-eu.researchsquare.com/files/rs-8307666/v1/c21820af0e22101cd650de42.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"SAGE: Decoupling Spatial Logic from Metric Scale for Zero-Shot Multi-Robot Exploration","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8307666/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8307666/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Effective multi-robot exploration in unknown environments faces a critical structural dilemma: high-level coordination typically relies on global spatial priors or fixed-scale representations, yet real-world missions operate in unbounded, irregular domains where such priors are unavailable. Here, we present SAGE, a framework that resolves this dilemma by treating exploration as a topological invariant, independent of metric boundaries. SAGE acts as a dimensional reduction operator, abstracting high-dimensional sensor streams into a sparse, dynamic graph that decouples decision logic from physical scale. To stabilize learning on these continuously growing structures, we introduce a capacity-controlled multi-agent reinforcement learning regime. By synchronizing the agent's observable horizon with its effective action space via a dynamic masking protocol, this mechanism constrains the complexity of the growing environment, ensuring the policy masters local topological correlations. We demonstrate that policies trained exclusively on minimalist, structured 60 × 60 grid worlds transfer zero-shot to environments of vastly different physics and scales—from large-scale unstructured maps to real-world deployment in dense, irregular forests. These results verify that abstracting spatial logic from metric geometry enables robust, boundless exploration without environment-specific tuning.","manuscriptTitle":"SAGE: Decoupling Spatial Logic from Metric Scale for Zero-Shot Multi-Robot Exploration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-16 08:21:19","doi":"10.21203/rs.3.rs-8307666/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d9e2bf9d-7141-4550-a967-00d025f06abb","owner":[],"postedDate":"December 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":59679744,"name":"Physical sciences/Engineering/Mechanical engineering"},{"id":59679745,"name":"Physical sciences/Engineering/Aerospace engineering"}],"tags":[],"updatedAt":"2026-03-12T10:46:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-16 08:21:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8307666","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8307666","identity":"rs-8307666","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-05-24T02:00:01.246996+00:00
License: CC-BY-4.0