{"paper_id":"4e244750-44a8-4efb-b409-d557ce106f0c","body_text":"Navigation in a Three-Dimensional Urban Flow using Deep Reinforcement Learning | 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 Navigation in a Three-Dimensional Urban Flow using Deep Reinforcement Learning Federica Tonti, Ricardo Vinuesa This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7249313/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 Unmanned Aerial Vehicles (UAVs) are increasingly populating urban areas for delivery and surveillance purposes. In this work, we develop an optimal navigation strategy based on Deep Reinforcement Learning. The environment is represented by a three-dimensional high-fidelity simulation of an urban flow, characterized by turbulence and recirculation zones. The algorithm presented here is a flow-aware Proximal Policy Optimization (PPO) combined with a Gated Transformer eXtra Large (GTrXL) architecture, giving the agent richer information about the turbulent flow field in which it navigates. The results are compared with a PPO+GTrXL without the secondary prediction tasks, a PPO combined with Long Short Term Memory (LSTM) cells and a traditional navigation algorithm. The obtained results show a significant increase in the success rate (SR) and a lower crash rate (CR) compared to a PPO+LSTM, PPO+GTrXL and the classical Zermelo's navigation algorithm, paving the way to a completely reimagined UAV landscape in complex urban environments. Physical sciences/Engineering/Aerospace engineering Physical sciences/Mathematics and computing/Computational science Deep Reinforcement Learning UAV Navigation Urban environment Full Text Additional Declarations There is NO Competing Interest. 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-7249313\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":503062507,\"identity\":\"cec913b9-0cd0-42b7-8914-f78a456d9d02\",\"order_by\":0,\"name\":\"Federica Tonti\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAoklEQVRIiWNgGAWjYBAC9gYQaWBjwCBBrBaeY0DigEGaAQ+JWhgOk6JFvvnY4w8F543tpRuYP3wgSgsbW7rBAYPbZjwyB9gkZxCjxZ6Nx0wCqMWGRyKBjZmHOFvAWs6BtDB//kOClgNmQC0M0sToAGpJS5M4Y5BszHMjsU2yhygtzIePSVT8sTNsn5F8+MMPoqxBAMYGEjWMglEwCkbBKMAJAMJdKmuttNpOAAAAAElFTkSuQmCC\",\"orcid\":\"\",\"institution\":\"KTH\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Federica\",\"middleName\":\"\",\"lastName\":\"Tonti\",\"suffix\":\"\"},{\"id\":503062508,\"identity\":\"72e1398d-3ba1-4b9b-8f0f-d32ebdba11a1\",\"order_by\":1,\"name\":\"Ricardo Vinuesa\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-6570-5499\",\"institution\":\"KTH Royal Institute of Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Ricardo\",\"middleName\":\"\",\"lastName\":\"Vinuesa\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-07-30 06:40:56\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-7249313/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-7249313/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":94826488,\"identity\":\"012d00b5-b79d-444e-a86e-9a8b149140c7\",\"added_by\":\"auto\",\"created_at\":\"2025-10-31 06:51:52\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":3102259,\"visible\":true,\"origin\":\"\",\"legend\":\"Article File\",\"description\":\"\",\"filename\":\"Navigation3DTonti.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7249313/v1_covered_0e2f20fe-0772-4461-a2bb-a221dacc771a.pdf\"}],\"financialInterests\":\"There is \\u003cb\\u003eNO\\u003c/b\\u003e Competing Interest.\",\"formattedTitle\":\"Navigation in a Three-Dimensional Urban Flow\\nusing Deep Reinforcement Learning\",\"fulltext\":[],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":false,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"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\":\"info@researchsquare.com\",\"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\":\"Deep Reinforcement Learning, UAV, Navigation, Urban environment\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-7249313/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-7249313/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"Unmanned Aerial Vehicles (UAVs) are increasingly populating urban areas for delivery and surveillance purposes. In this work, we develop an optimal navigation strategy based on Deep Reinforcement Learning. The environment is represented by a three-dimensional high-fidelity simulation of an urban flow, characterized by turbulence and recirculation zones. The algorithm presented here is a flow-aware Proximal Policy Optimization (PPO) combined with a Gated Transformer eXtra Large (GTrXL) architecture, giving the agent richer information about the turbulent flow field in which it navigates. The results are compared with a PPO+GTrXL without the secondary prediction tasks, a PPO combined with Long Short Term Memory (LSTM) cells and a traditional navigation algorithm. The obtained results show a significant increase in the success rate (SR) and a lower crash rate (CR) compared to a PPO+LSTM, PPO+GTrXL and the classical Zermelo's navigation algorithm, paving the way to a completely reimagined UAV landscape in complex urban environments.\",\"manuscriptTitle\":\"Navigation in a Three-Dimensional Urban Flow\\nusing Deep Reinforcement Learning\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-10-31 03:24:25\",\"doi\":\"10.21203/rs.3.rs-7249313/v1\",\"editorialEvents\":[],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"communications-engineering\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"commseng\",\"sideBox\":\"Learn more about [Communications Engineering](http://link.springer.com/journal/44172)\",\"snPcode\":\"44172\",\"submissionUrl\":\"https://mts-commseng.nature.com/cgi-bin/main.plex\",\"title\":\"Communications Engineering\",\"twitterHandle\":\"@commseng\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"Springer Hybrid\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"d1d27778-9b0c-4132-8e55-52fbc16cd4d4\",\"owner\":[],\"postedDate\":\"October 31st, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"under-review\",\"subjectAreas\":[{\"id\":53441077,\"name\":\"Physical sciences/Engineering/Aerospace engineering\"},{\"id\":53441078,\"name\":\"Physical sciences/Mathematics and computing/Computational science\"}],\"tags\":[],\"updatedAt\":\"2025-10-31T03:24:25+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2025-10-31 03:24:25\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-7249313\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-7249313\",\"identity\":\"rs-7249313\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}