Architecture-independent analysis of task conflicts in data-driven end-to-end controllers for lower-limb wearable robots

Architecture-independent analysis of task conflicts in data-driven end-to-end controllers for lower-limb wearable robots | 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 Architecture-independent analysis of task conflicts in data-driven end-to-end controllers for lower-limb wearable robots José A. Montes Pérez, Ryan Posh, Gray Thomas, Robert D. Gregg This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7668449/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 End-to-end control strategies for lower-limb wearable robots map sensor data directly to joint moment predictions, generalizing assistance across locomotor tasks without explicitly classifying them. These control models are iteratively trained/validated with varied sensor inputs and time windows, which is a costly process that conflates the performance of the model architecture with fundamen- tal conflicts in the data. Here we present an architecture-independent framework that quantifies input-output conflicts across tasks using multivariate Gaussian models of phase-dependent biomechanical data. Conflict heatmaps reveal gait phases where similar sensor inputs demand contradictory torque outputs between tasks. Analysis of the hip, knee, and ankle shows the relative efficacy of different input sources and time windows at reducing conflicts. We also find an empirical correlation between input-output conflict and model error for example archi- tectures. Supported by an online tool, this framework enables principled sensor selection and conflict analysis to advance the development of versatile prostheses and exoskeletons. Physical sciences/Mathematics and computing/Applied mathematics Physical sciences/Engineering/Mechanical engineering Physical sciences/Engineering/Electrical and electronic engineering Physical sciences/Engineering/Biomedical engineering Exoskeletons and Prostheses Learning-Based Control Sensor Selection Full Text Additional Declarations There is NO Competing Interest. 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-7668449","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":543701191,"identity":"7df69b57-ab04-4d16-9cc8-57e16ba1f0b8","order_by":0,"name":"José A. Montes Pérez","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsklEQVRIiWNgGAWjYBACxgYwZcPAwMzAxgAiidNygCGNBC1gcIDhMIgiUgvztMPHHn/ccT5xbTvzswcMFdaJDQQdNjst3eDgmduJ2w6zmRswnEknRkuOmcTBNpAWHjYJxrbDRGs5B9Xyj3gtB6BaGojSkpYmcfZMsjHQL2YSCcfSjQlqMZydfEyicoed7Lbzh59JfKixliWsBaSCEaYsgZByEJBnQNYyCkbBKBgFowAbAAC+5EEkniQ4IAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0009-0001-8873-3254","institution":"Robotics Department, University of Michigan, 2505 Hayward St., Ann Arbor, 48109, Michigan, USA","correspondingAuthor":true,"prefix":"","firstName":"José","middleName":"A. Montes","lastName":"Pérez","suffix":""},{"id":543701193,"identity":"5e9ce764-f866-445c-b41f-837e750143cb","order_by":1,"name":"Ryan Posh","email":"","orcid":"","institution":"Robotics Department, University of Michigan, 2505 Hayward St., Ann Arbor, 48109, Michigan, USA","correspondingAuthor":false,"prefix":"","firstName":"Ryan","middleName":"","lastName":"Posh","suffix":""},{"id":543701194,"identity":"396a2798-35b4-4bd4-aa01-2708941b8637","order_by":2,"name":"Gray Thomas","email":"","orcid":"","institution":"Department of Mechanical Engineering, Texas A\u0026M University, 3123 TAMU, College Station, 77843, Texas, USA","correspondingAuthor":false,"prefix":"","firstName":"Gray","middleName":"","lastName":"Thomas","suffix":""},{"id":543701192,"identity":"c0622663-0594-4782-b1a7-c7ee9c9eaa2d","order_by":3,"name":"Robert D. Gregg","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxElEQVRIiWNgGAWjYFCCBDApx8DA2MDMACKJ1WJMupZEkEritJi3p1/78OFXbfqGa4cbPxcw2MhuOEBAi8yZN8UzZ/Ydz91wO7FZegZDmjFBLRISOcnMvD3HQFramHkYDicSp+Vvz7F0A4iW/8RoST/MzPCjJgGq5QARWnjeMDP2NhwwnAnyC49BsvFMglrY0x8z/PhTJ893O/3hZ54KO9k+QloYGHgMGBjbDkM5BgSVgwD7AwaGP3VEKR0Fo2AUjIIRCgBmyUg3Hh2l6gAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-0729-2857","institution":"Robotics Department, University of Michigan, 2505 Hayward St., Ann Arbor, 48109, Michigan, USA","correspondingAuthor":true,"prefix":"","firstName":"Robert","middleName":"D.","lastName":"Gregg","suffix":""}],"badges":[],"createdAt":"2025-09-21 20:10:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7668449/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7668449/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97896412,"identity":"a409f7c2-eceb-4fb1-a698-3d394a618723","added_by":"auto","created_at":"2025-12-10 15:36:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4224682,"visible":true,"origin":"","legend":"Article File","description":"","filename":"Architectureindependentanalysisoftaskconflictsforendtoendlowerlimbrobots.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7668449/v1_covered_8eb8e61c-ecd1-401d-b880-e10d1573eac0.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Architecture-independent analysis of task conflicts in data-driven end-to-end controllers for lower-limb wearable robots","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"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":"Exoskeletons and Prostheses, Learning-Based Control, Sensor Selection","lastPublishedDoi":"10.21203/rs.3.rs-7668449/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7668449/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"End-to-end control strategies for lower-limb wearable robots map sensor data\r\ndirectly to joint moment predictions, generalizing assistance across locomotor\r\ntasks without explicitly classifying them. These control models are iteratively\r\ntrained/validated with varied sensor inputs and time windows, which is a costly\r\nprocess that conflates the performance of the model architecture with fundamen-\r\ntal conflicts in the data. Here we present an architecture-independent framework\r\nthat quantifies input-output conflicts across tasks using multivariate Gaussian\r\nmodels of phase-dependent biomechanical data. Conflict heatmaps reveal gait\r\nphases where similar sensor inputs demand contradictory torque outputs between\r\ntasks. Analysis of the hip, knee, and ankle shows the relative efficacy of different\r\ninput sources and time windows at reducing conflicts. We also find an empirical\r\ncorrelation between input-output conflict and model error for example archi-\r\ntectures. Supported by an online tool, this framework enables principled sensor\r\nselection and conflict analysis to advance the development of versatile prostheses\r\nand exoskeletons.","manuscriptTitle":"Architecture-independent analysis of task conflicts in data-driven end-to-end controllers for lower-limb wearable robots","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-13 04:19:23","doi":"10.21203/rs.3.rs-7668449/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":"69009505-803b-4ef4-b9e7-1de475216412","owner":[],"postedDate":"November 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":57844151,"name":"Physical sciences/Mathematics and computing/Applied mathematics"},{"id":57844152,"name":"Physical sciences/Engineering/Mechanical engineering"},{"id":57844153,"name":"Physical sciences/Engineering/Electrical and electronic engineering"},{"id":57844154,"name":"Physical sciences/Engineering/Biomedical engineering"}],"tags":[],"updatedAt":"2026-03-23T13:21:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-13 04:19:23","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7668449","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7668449","identity":"rs-7668449","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}