{"paper":{"title":"From Gridworlds to Warehouses: Adapting Lightweight One-shot Multi-Agent Pathfinding for AGVs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Adapting MAPF for warehouse AGVs with motion constraints shows PIBT scales better than PP or LNS2 for large teams.","cross_cats":["cs.RO"],"primary_cat":"cs.MA","authors_text":"Hiroki Nagai, Keisuke Okumura","submitted_at":"2026-05-15T09:55:35Z","abstract_excerpt":"Multi-agent pathfinding (MAPF) under one-shot planning is a core component of warehouse automation, yet classical formulations typically assume four-connected 2D grids with unit-time moves in four directions. To fill reality gaps while still being trackable with discrete combinatorial search, this work proposes a more practical counterpart tailored to differential-drive AGVs. We term this multi-agent warehouse pathfinding (MAWPF), featured with four constraints: (i) agent actions are restricted to straight motion and in-place rotation; (ii) rotations require multi-step costs; (iii) acceleratio"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our experiments reveal that PP and LNS2 struggle to solve instances with many agents, while PIBT-based approaches achieve preferable scalability with increased solution cost.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The four listed constraints (straight motion only, multi-step rotations, acceleration/deceleration, and follower-collision prohibition) are assumed to be sufficient to close the main reality gaps for differential-drive AGVs while remaining tractable for discrete search.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Adapts PP, LNS2, PIBT and LaCAM to MAWPF with AGV motion constraints and shows PIBT variants scale better than PP or LNS2 at the cost of longer paths.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Adapting MAPF for warehouse AGVs with motion constraints shows PIBT scales better than PP or LNS2 for large teams.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"214e6dcac31d07a271eefef0844663d1b115d9312271c227367108479d1d76ab"},"source":{"id":"2605.15799","kind":"arxiv","version":1},"verdict":{"id":"6491181a-4cc5-4d7d-a09c-fee53ae75412","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T18:54:20.877118Z","strongest_claim":"Our experiments reveal that PP and LNS2 struggle to solve instances with many agents, while PIBT-based approaches achieve preferable scalability with increased solution cost.","one_line_summary":"Adapts PP, LNS2, PIBT and LaCAM to MAWPF with AGV motion constraints and shows PIBT variants scale better than PP or LNS2 at the cost of longer paths.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The four listed constraints (straight motion only, multi-step rotations, acceleration/deceleration, and follower-collision prohibition) are assumed to be sufficient to close the main reality gaps for differential-drive AGVs while remaining tractable for discrete search.","pith_extraction_headline":"Adapting MAPF for warehouse AGVs with motion constraints shows PIBT scales better than PP or LNS2 for large teams."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15799/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T19:01:37.572379Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T19:01:19.016873Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:48.739400Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:21:55.903004Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"c3c02c809d566c139173be3d09367fb232db6e9778104e8230bea41791f1fe0c"},"references":{"count":21,"sample":[{"doi":"","year":2021,"title":"Prioritized sipp for multi-agent path finding with kinematic constraints","work_id":"607124bc-df15-4b13-b7c8-39cfe2b4565f","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"The league of robot runners competition: Goals, designs, and implementation","work_id":"b4c384cc-d293-4ae1-87c2-c1962d0d9d55","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2008,"title":"A multiagent approach to autonomous intersection manage- ment.Journal of Artificial Intelligence Research (JAIR),","work_id":"c1d09c4f-f5c6-480c-a28d-3588b885ac72","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1987,"title":"On multiple moving objects.Al- gorithmica,","work_id":"a81fd6ae-18c9-443d-ad31-78994c88831b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1968,"title":"[Hartet al., 1968 ] Peter E. Hart, Nils J. Nilsson, and Bertram Raphael. A formal basis for the heuristic determination of minimum cost paths.IEEE Transactions on Systems Science and Cybernetics,","work_id":"50fcf795-a45b-4b46-ae0b-0bbba030f6c1","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":21,"snapshot_sha256":"1bed11826791ad476b3ab9083518cc531d55ec8b2d9ce19838e02feb2886e666","internal_anchors":1},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}