NeRF-Nav: Hierarchical Neural Radiance Fields for Real-Time Robot Navigation and Obstacle Avoidance

NeRF-Nav: Hierarchical Neural Radiance Fields for Real-Time Robot Navigation and Obstacle Avoidance | 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 Research Article NeRF-Nav: Hierarchical Neural Radiance Fields for Real-Time Robot Navigation and Obstacle Avoidance Kenji Takahashi, Ayumi Watanabe, Jayden Mercer, Sho Nakamura, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8887612/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Neural radiance fields (NeRFs) offer photorealistic scene representations but their monolithic structure and slow rendering hinder deployment for real-time robot navigation. We present NeRF-Nav, a hierarchical NeRF framework that enables real-time obstacle avoidance and path planning by decomposing large environments into a tree of local radiance fields with varying levels of detail. Our system introduces: (i) an occupancy-aware NeRF variant that jointly learns density and a binary occupancy grid for collision checking in constant time, (ii) a hierarchical allocation strategy that spawns and prunes local NeRF nodes based on the robot's exploration frontier, and (iii) a neural potential field planner that extracts repulsive gradients directly from the radiance field density without explicit mesh extraction. Evaluated on the Gibson, Matterport3D, and a custom warehouse dataset, NeRF-Nav achieves 94.6% collision-free navigation success at 18 Hz planning rate, outperforming both voxel-grid and TSDF baselines by 8–15% in cluttered environments. Our approach reduces memory usage by 3.8x compared to a single global NeRF while maintaining rendering quality (PSNR within 0.3 dB). Robotics Neural radiance fields robot navigation obstacle avoidance path planning hierarchical scene representation occupancy mapping Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted 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-8887612","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":591695044,"identity":"5fe35eca-80b2-48c1-b641-ce76050e7ae8","order_by":0,"name":"Kenji Takahashi","email":"","orcid":"","institution":"Nara Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Kenji","middleName":"","lastName":"Takahashi","suffix":""},{"id":591695045,"identity":"7cc0e73c-07ee-4583-80b7-393e54b1962f","order_by":1,"name":"Ayumi Watanabe","email":"","orcid":"","institution":"Nara Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Ayumi","middleName":"","lastName":"Watanabe","suffix":""},{"id":591695046,"identity":"795d07a7-4888-4cee-85d0-428ee12b1b43","order_by":2,"name":"Jayden Mercer","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYPACNsZ+MG1gAeHzEKNlZgNYiwTRWhgYNxwA00Ro0W3vffjh4x4+2c03kp9u+FEgwWBw/ADjg7dtuLWYnTluLDnjGZvxthtpZjd7gA4zOJPAbDgXn5YbaWzMPAfYErfdyGG7wQPScoOBTZoXn5b7z9iY/wC1bJ6Rw3bzD0QL+2+8Wm6wsTEzALVskMhhuw2zhRmvljNpzJI9B9iMZ5x5ZnZbxkCCR/JMYrPknHN4tBw/xvjhx4Fjsv3tyc9uvvljI8d3/PDBD2/KcGuBgmNwFjBGGBsIqgeCGmIUjYJRMApGwUgFAJ+3UEyRiuzOAAAAAElFTkSuQmCC","orcid":"","institution":"University of Cincinnati","correspondingAuthor":true,"prefix":"","firstName":"Jayden","middleName":"","lastName":"Mercer","suffix":""},{"id":591695047,"identity":"c0ab82c7-d464-48b5-b49c-10f0ce3b96e7","order_by":3,"name":"Sho Nakamura","email":"","orcid":"","institution":"Osaka Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Sho","middleName":"","lastName":"Nakamura","suffix":""},{"id":591695048,"identity":"be9160a2-7618-4c46-9aa4-5780f0af346b","order_by":4,"name":"Haruki Yamamoto","email":"","orcid":"","institution":"Nara Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Haruki","middleName":"","lastName":"Yamamoto","suffix":""}],"badges":[],"createdAt":"2026-02-15 17:28:34","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8887612/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8887612/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102963547,"identity":"630c3921-641d-4051-89c2-c2bb53dfe368","added_by":"auto","created_at":"2026-02-19 04:18:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":223645,"visible":true,"origin":"","legend":"","description":"","filename":"nerfNAV.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8887612/v1_covered_a1f953b7-f63f-4997-8ced-5baceff05462.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eNeRF-Nav: Hierarchical Neural Radiance Fields for Real-Time Robot Navigation and Obstacle Avoidance\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[{"identity":"39b81bcd-78d4-4749-970a-8ebc675117bd","identifier":"10.13039/501100001691","name":"Japan Society for the Promotion of Science","awardNumber":"JP23K13456","order_by":0},{"identity":"85876ba4-d16e-4dbf-a517-c09b4d689c4d","identifier":"10.13039/501100002241","name":"Japan Science and Technology Agency","awardNumber":"JPMJCR20D6","order_by":1},{"identity":"5b82e181-7750-42ca-9673-e5c0b6f47f8b","identifier":"10.13039/100000001","name":"National Science Foundation","awardNumber":"OISE-2230283","order_by":2}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Nara Institute of Science and Technology","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Neural radiance fields, robot navigation, obstacle avoidance, path planning, hierarchical scene representation, occupancy mapping","lastPublishedDoi":"10.21203/rs.3.rs-8887612/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8887612/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eNeural radiance fields (NeRFs) offer photorealistic scene representations but their monolithic structure and slow rendering hinder deployment for real-time robot navigation. We present NeRF-Nav, a hierarchical NeRF framework that enables real-time obstacle avoidance and path planning by decomposing large environments into a tree of local radiance fields with varying levels of detail. Our system introduces: (i) an occupancy-aware NeRF variant that jointly learns density and a binary occupancy grid for collision checking in constant time, (ii) a hierarchical allocation strategy that spawns and prunes local NeRF nodes based on the robot's exploration frontier, and (iii) a neural potential field planner that extracts repulsive gradients directly from the radiance field density without explicit mesh extraction. Evaluated on the Gibson, Matterport3D, and a custom warehouse dataset, NeRF-Nav achieves 94.6% collision-free navigation success at 18 Hz planning rate, outperforming both voxel-grid and TSDF baselines by 8–15% in cluttered environments. Our approach reduces memory usage by 3.8x compared to a single global NeRF while maintaining rendering quality (PSNR within 0.3 dB).\u003c/p\u003e","manuscriptTitle":"NeRF-Nav: Hierarchical Neural Radiance Fields for Real-Time Robot Navigation and Obstacle Avoidance","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-18 02:42:21","doi":"10.21203/rs.3.rs-8887612/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"74f8b022-3a03-49d4-afa1-693e1ea730a3","owner":[],"postedDate":"February 18th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":63102370,"name":"Robotics"}],"tags":[],"updatedAt":"2026-02-18T02:42:21+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-18 02:42:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8887612","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8887612","identity":"rs-8887612","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}