Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387156920> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4387156920 abstract "Terrain awareness, i.e., the ability to identify and distinguish different types of terrain, is a critical ability that robots must have to succeed at autonomous off-road navigation. Current approaches that provide robots with this awareness either rely on labeled data which is expensive to collect, engineered features and cost functions that may not generalize, or expert human demonstrations which may not be available. Towards endowing robots with terrain awareness without these limitations, we introduce Self-supervised TErrain Representation LearnING (STERLING), a novel approach for learning terrain representations that relies solely on easy-to-collect, unconstrained (e.g., non-expert), and unlabelled robot experience, with no additional constraints on data collection. STERLING employs a novel multi-modal self-supervision objective through non-contrastive representation learning to learn relevant terrain representations for terrain-aware navigation. Through physical robot experiments in off-road environments, we evaluate STERLING features on the task of preference-aligned visual navigation and find that STERLING features perform on par with fully supervised approaches and outperform other state-of-the-art methods with respect to preference alignment. Additionally, we perform a large-scale experiment of autonomously hiking a 3-mile long trail which STERLING completes successfully with only two manual interventions, demonstrating its robustness to real-world off-road conditions." @default.
- W4387156920 created "2023-09-30" @default.
- W4387156920 creator A5001594330 @default.
- W4387156920 creator A5004302220 @default.
- W4387156920 creator A5039434686 @default.
- W4387156920 creator A5060981901 @default.
- W4387156920 creator A5062572089 @default.
- W4387156920 creator A5092964245 @default.
- W4387156920 date "2023-09-26" @default.
- W4387156920 modified "2023-09-30" @default.
- W4387156920 title "Self-Supervised Terrain Representation Learning from Unconstrained Robot Experience" @default.
- W4387156920 doi "https://doi.org/10.48550/arxiv.2309.15302" @default.
- W4387156920 hasPublicationYear "2023" @default.
- W4387156920 type Work @default.
- W4387156920 citedByCount "0" @default.
- W4387156920 crossrefType "posted-content" @default.
- W4387156920 hasAuthorship W4387156920A5001594330 @default.
- W4387156920 hasAuthorship W4387156920A5004302220 @default.
- W4387156920 hasAuthorship W4387156920A5039434686 @default.
- W4387156920 hasAuthorship W4387156920A5060981901 @default.
- W4387156920 hasAuthorship W4387156920A5062572089 @default.
- W4387156920 hasAuthorship W4387156920A5092964245 @default.
- W4387156920 hasBestOaLocation W43871569201 @default.
- W4387156920 hasConcept C104317684 @default.
- W4387156920 hasConcept C107457646 @default.
- W4387156920 hasConcept C119857082 @default.
- W4387156920 hasConcept C154945302 @default.
- W4387156920 hasConcept C161840515 @default.
- W4387156920 hasConcept C17744445 @default.
- W4387156920 hasConcept C185592680 @default.
- W4387156920 hasConcept C199539241 @default.
- W4387156920 hasConcept C205649164 @default.
- W4387156920 hasConcept C2776359362 @default.
- W4387156920 hasConcept C41008148 @default.
- W4387156920 hasConcept C55493867 @default.
- W4387156920 hasConcept C58640448 @default.
- W4387156920 hasConcept C59404180 @default.
- W4387156920 hasConcept C63479239 @default.
- W4387156920 hasConcept C90509273 @default.
- W4387156920 hasConcept C94625758 @default.
- W4387156920 hasConceptScore W4387156920C104317684 @default.
- W4387156920 hasConceptScore W4387156920C107457646 @default.
- W4387156920 hasConceptScore W4387156920C119857082 @default.
- W4387156920 hasConceptScore W4387156920C154945302 @default.
- W4387156920 hasConceptScore W4387156920C161840515 @default.
- W4387156920 hasConceptScore W4387156920C17744445 @default.
- W4387156920 hasConceptScore W4387156920C185592680 @default.
- W4387156920 hasConceptScore W4387156920C199539241 @default.
- W4387156920 hasConceptScore W4387156920C205649164 @default.
- W4387156920 hasConceptScore W4387156920C2776359362 @default.
- W4387156920 hasConceptScore W4387156920C41008148 @default.
- W4387156920 hasConceptScore W4387156920C55493867 @default.
- W4387156920 hasConceptScore W4387156920C58640448 @default.
- W4387156920 hasConceptScore W4387156920C59404180 @default.
- W4387156920 hasConceptScore W4387156920C63479239 @default.
- W4387156920 hasConceptScore W4387156920C90509273 @default.
- W4387156920 hasConceptScore W4387156920C94625758 @default.
- W4387156920 hasLocation W43871569201 @default.
- W4387156920 hasOpenAccess W4387156920 @default.
- W4387156920 hasPrimaryLocation W43871569201 @default.
- W4387156920 hasRelatedWork W1521672401 @default.
- W4387156920 hasRelatedWork W1989587773 @default.
- W4387156920 hasRelatedWork W2107591285 @default.
- W4387156920 hasRelatedWork W2891961174 @default.
- W4387156920 hasRelatedWork W3087493185 @default.
- W4387156920 hasRelatedWork W3213331859 @default.
- W4387156920 hasRelatedWork W4206762304 @default.
- W4387156920 hasRelatedWork W4226458444 @default.
- W4387156920 hasRelatedWork W4292217789 @default.
- W4387156920 hasRelatedWork W4383097398 @default.
- W4387156920 isParatext "false" @default.
- W4387156920 isRetracted "false" @default.
- W4387156920 workType "article" @default.