{"paper":{"title":"NORM-Nav: Zero-Shot Mobile Robot Navigation with Natural Language Behavioral Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"NORM-Nav converts natural language behavioral constraints into multi-layer costmaps that standard planners can use for more human-like robot paths.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chao Gao, Dongjie Huo, Dong Zhang, Guyue Zhou, Junhui Wang, Yan Qiao","submitted_at":"2026-05-16T12:58:03Z","abstract_excerpt":"Mobile robots operating in human-centered environments must generate not only collision-free paths but also trajectories that follow local behavioral conventions. Conventional costmap-based navigation emphasizes geometric feasibility and often overlooks such requirements, which can result in socially inappropriate behaviors. This paper presents NORM-Nav, a zero-shot framework that integrates natural language behavioral constraints into costmap-based planning. An LLM parses each instruction into structured constraints and grounds them using real-time vision--LiDAR perception. These constraints "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"NORM-Nav improves task success rates and produces trajectories closer to human references than representative baselines in simulation and real-world experiments.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The LLM can reliably parse natural language behavioral constraints and the real-time vision-LiDAR system can accurately ground those constraints to produce effective multi-layer costmaps that preserve intended behavior when used by standard planners.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NORM-Nav is a zero-shot framework that parses natural language behavioral constraints with an LLM, grounds them via vision-LiDAR, and encodes them as multi-layer costmaps for grid-based robot navigation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"NORM-Nav converts natural language behavioral constraints into multi-layer costmaps that standard planners can use for more human-like robot paths.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f96e252af9f1e73cc2674257c75360d5403de013d9dbf3e4bf09404984c44f13"},"source":{"id":"2605.16979","kind":"arxiv","version":1},"verdict":{"id":"ebd677d1-9f15-4a56-935c-82cacf448fc9","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T20:08:19.481910Z","strongest_claim":"NORM-Nav improves task success rates and produces trajectories closer to human references than representative baselines in simulation and real-world experiments.","one_line_summary":"NORM-Nav is a zero-shot framework that parses natural language behavioral constraints with an LLM, grounds them via vision-LiDAR, and encodes them as multi-layer costmaps for grid-based robot navigation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The LLM can reliably parse natural language behavioral constraints and the real-time vision-LiDAR system can accurately ground those constraints to produce effective multi-layer costmaps that preserve intended behavior when used by standard planners.","pith_extraction_headline":"NORM-Nav converts natural language behavioral constraints into multi-layer costmaps that standard planners can use for more human-like robot paths."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16979/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T20:31:19.045364Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:21:39.814099Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T19:50:06.731328Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"cited_work_retraction","ran_at":"2026-05-19T19:24:11.876772Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T18:41:56.216527Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.303921Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"08668532a5511605eafe5d194b259aed587c5b6b75207daff0b9e75ef69f7d10"},"references":{"count":32,"sample":[{"doi":"","year":2024,"title":"Safe-vln: Collision avoidance for vision-and-language navigation of autonomous robots operating in continuous environments,","work_id":"a8a78a05-e480-4ecd-8ca8-e54144cb4f4e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"MPC-DS: A safe path track- ing method for agvs in dynamic environments with dense obstacles,","work_id":"2aae3fe6-202b-4575-b96b-a172ce480c1c","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Openbench: A new benchmark and baseline for semantic navigation in smart logistics,","work_id":"3296388c-52ac-4748-a673-543d757917a7","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Open: Lightweight map-based semantic navigation for gps-free last-mile delivery,","work_id":"bdceac25-51d9-482a-954d-45f140632f0e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"The marathon 2: A navigation system,","work_id":"7a652005-e7fc-4ed9-a4b6-aade39e8fd38","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":32,"snapshot_sha256":"51bd69456af0489c644d39161a6341e7634ec5ac32f8c0d7b17b091fa5656539","internal_anchors":4},"formal_canon":{"evidence_count":2,"snapshot_sha256":"31ba05e12abb699f50ea892ac7b2fbbbc63290b44ffd78d3972c1b40dceecfdd"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}