A Dual‑Agent Learning Framework Enables Level 3+ Autonomous Robotic Bronchoscopy | 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 A Dual‑Agent Learning Framework Enables Level 3+ Autonomous Robotic Bronchoscopy Hongen Liao, Guochen Ning, Yiguang Yang, Jiayuan Liu, Jiajun Ma, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8414938/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 Accuracy access to peripheral lung lesions is critical for diagnosis and treatment, yet clinical bronchoscopy remains limited by operator variability and loss of spatial accuracy during respiratory motion. Conventional navigation systems depend on preoperative maps and external tracking, which fail in the deformable and continuously moving bronchial tree. We developed a fully autonomous, adaptive bronchoscopy platform that performs real‑time navigation directly from intraluminal visual-spatial cues, without predefined trajectories or external localization. A hierarchical dual‑agent framework enables a Navigator to infer dynamic waypoints and a Driver to execute fine‑scale steering with closed‑loop precision. In dynamically ventilated ex vivo lungs and live canine models, the system achieved consistent, lesion‑targeted navigation under rapid respiration without human input. Beyond bronchoscopy, this approach defines a general computational paradigm for autonomy in deformable living anatomy, offering a clinically translatable path toward operator‑independent intervention across organ systems. Health sciences/Diseases/Respiratory tract diseases Physical sciences/Engineering/Biomedical engineering Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SuppMovie1.mp4 Ex-vivo experiment SuppMovie2.mp4 in-vivo experiment 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-8414938","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":573548249,"identity":"3a0e12b3-fc5c-4dc6-a099-c11fe59d8567","order_by":0,"name":"Hongen Liao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4ElEQVRIiWNgGAWjYBACPmYg8YGBgQeMgICxgZAWNqAWxhkJJGkBYmaeBBCTaC3szA8f2/6okzHnX3tM4geDjeyGA8zPHuB3GJuxcU7CYR7LGe/SJHsY0ow3HGAzNyDgFzPpnIQDPAY3zphJ8DAcTtxwgIdNAr8W9m/SFgl1YC2Sfxj+E6OFx0yaIYGZx+B8j5k0D8MBorQUG/akHQbawmNsLWOQbDzzMJsZXi38/Mc3PvhhU2dvcP6M4c03FXayfcebn+HVggASCUACFFTMxKkH2XeAaKWjYBSMglEwwgAAiDY+EeSPqxQAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0003-3847-9347","institution":"Tsinghua University","correspondingAuthor":true,"prefix":"","firstName":"Hongen","middleName":"","lastName":"Liao","suffix":""},{"id":573548250,"identity":"9e14165b-f5cf-4700-8c84-c9af659ef056","order_by":1,"name":"Guochen Ning","email":"","orcid":"https://orcid.org/0000-0003-0282-7303","institution":"Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China","correspondingAuthor":false,"prefix":"","firstName":"Guochen","middleName":"","lastName":"Ning","suffix":""},{"id":573548251,"identity":"d8c88cdf-64f9-4a80-97b0-56da1bf3145e","order_by":2,"name":"Yiguang Yang","email":"","orcid":"","institution":"School of Biomedical Engineering, Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Yiguang","middleName":"","lastName":"Yang","suffix":""},{"id":573548252,"identity":"7ffd7cf6-8369-40f7-9893-99e1983f9c3e","order_by":3,"name":"Jiayuan Liu","email":"","orcid":"","institution":"School of Biomedical Engineering, Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Jiayuan","middleName":"","lastName":"Liu","suffix":""},{"id":573548253,"identity":"c08b9cab-ff2d-4a13-aa14-eb8e5cfc3feb","order_by":4,"name":"Jiajun Ma","email":"","orcid":"","institution":"School of Biomedical Engineering, Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Jiajun","middleName":"","lastName":"Ma","suffix":""},{"id":573548254,"identity":"29eaef8f-3502-452d-b1b4-2a48a4c7d567","order_by":5,"name":"Fang Chen","email":"","orcid":"","institution":"School of Biomedical Engineering, Shanghai Jiao Tong University","correspondingAuthor":false,"prefix":"","firstName":"Fang","middleName":"","lastName":"Chen","suffix":""},{"id":573548255,"identity":"58a357ba-cd79-4a89-9e41-97ffc89736c7","order_by":6,"name":"Zhe Zhao","email":"","orcid":"","institution":"School of Clinical Medicine (Beijing Tsinghua Changgung Hospital), Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Zhe","middleName":"","lastName":"Zhao","suffix":""},{"id":573548256,"identity":"489587f3-9118-4c4d-8310-31807c5c8d7f","order_by":7,"name":"Xiangdong Mu","email":"","orcid":"","institution":"School of Clinical Medicine (Beijing Tsinghua Changgung Hospital), Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Xiangdong","middleName":"","lastName":"Mu","suffix":""}],"badges":[],"createdAt":"2025-12-21 03:20:59","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8414938/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8414938/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102298634,"identity":"85ca1fd6-262f-4447-b0d4-96dc9cab635b","added_by":"auto","created_at":"2026-02-10 10:54:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1810495,"visible":true,"origin":"","legend":"","description":"","filename":"ADualXXXAgentLearningFrameworkEnablesLevel3AutonomousRoboticBronchoscopy.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8414938/v1_covered_de7af9ec-863f-4604-86db-21eb0ab5b5b5.pdf"},{"id":101967202,"identity":"c572563a-b346-4221-b0d5-a9182ceb7590","added_by":"auto","created_at":"2026-02-05 13:56:59","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":26076220,"visible":true,"origin":"","legend":"\u003cp\u003eEx-vivo experiment\u003c/p\u003e","description":"","filename":"SuppMovie1.mp4","url":"https://assets-eu.researchsquare.com/files/rs-8414938/v1/3d9ef6a4817a613971544b7e.mp4"},{"id":101967306,"identity":"4ce05643-dbc7-4527-b5a4-a2b179c0983a","added_by":"auto","created_at":"2026-02-05 13:57:18","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2621639,"visible":true,"origin":"","legend":"\u003cp\u003ein-vivo experiment\u003c/p\u003e","description":"","filename":"SuppMovie2.mp4","url":"https://assets-eu.researchsquare.com/files/rs-8414938/v1/ef8e2d22b60bc72d8ccab2ae.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"A Dual‑Agent Learning Framework Enables Level 3+ Autonomous Robotic Bronchoscopy","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-8414938/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8414938/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccuracy access to peripheral lung lesions is critical for diagnosis and treatment, yet clinical bronchoscopy remains limited by operator variability and loss of spatial accuracy during respiratory motion. Conventional navigation systems depend on preoperative maps and external tracking, which fail in the deformable and continuously moving bronchial tree. We developed a fully autonomous, adaptive bronchoscopy platform that performs real‑time navigation directly from intraluminal visual-spatial cues, without predefined trajectories or external localization. A hierarchical dual‑agent framework enables a Navigator to infer dynamic waypoints and a Driver to execute fine‑scale steering with closed‑loop precision. In dynamically ventilated ex vivo lungs and live canine models, the system achieved consistent, lesion‑targeted navigation under rapid respiration without human input. Beyond bronchoscopy, this approach defines a general computational paradigm for autonomy in deformable living anatomy, offering a clinically translatable path toward operator‑independent intervention across organ systems.\u003c/p\u003e","manuscriptTitle":"A Dual‑Agent Learning Framework Enables Level 3+ Autonomous Robotic Bronchoscopy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-05 13:56:36","doi":"10.21203/rs.3.rs-8414938/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":"c547d975-6746-4990-8283-d56163e65106","owner":[],"postedDate":"February 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":61026783,"name":"Health sciences/Diseases/Respiratory tract diseases"},{"id":61026784,"name":"Physical sciences/Engineering/Biomedical engineering"}],"tags":[],"updatedAt":"2026-02-05T13:56:36+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-05 13:56:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8414938","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8414938","identity":"rs-8414938","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","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.