{"paper":{"title":"Privacy Preserving Multi Agent Path Finding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Adding mock agents during planning lets multiple agents find collision-free paths without revealing their exact planned locations to each other.","cross_cats":[],"primary_cat":"cs.MA","authors_text":"Guy Shani, Roni Stern, Rotem Lev Lehman","submitted_at":"2026-05-13T21:08:24Z","abstract_excerpt":"In the multi-agent path finding (MAPF) problem, a group of agents search in a graph for a path for each agent where no two paths collide. While in all applications of MAPF the agents must not collide with each other, in some of them the agents may not wish to share their paths due to privacy constraints. In this work, we formulate two types of privacy constraints for MAPF and propose algorithms that preserve them. The first type of privacy we consider is planning-level privacy, which means that during planning, the agents cannot identify exactly the planned location of the other agents. We pro"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We propose a general framework for obtaining planning-level privacy, which works by adding mock agents to the planning process. We show how to adapt two popular MAPF algorithms, namely PIBT and LaCAM, such that they preserve execution-level privacy. Lastly, we propose a post-processing technique that allows the agents to reduce the sum of costs of the returned solution without losing any privacy.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that inserting mock agents during planning sufficiently obscures real agent locations for all participants without introducing detectable information leaks or violating collision constraints, and that the privacy-preserving modifications to PIBT and LaCAM retain the original algorithms' correctness properties.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"New algorithms for multi-agent path finding preserve planning-level privacy via mock agents and execution-level privacy via modified PIBT and LaCAM solvers, plus post-processing that lowers total cost without breaking privacy.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Adding mock agents during planning lets multiple agents find collision-free paths without revealing their exact planned locations to each other.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f3d4d015c0b8c0e4828fd05308b61827fa79814b181ca2a0f70ead1b104f3c69"},"source":{"id":"2605.14119","kind":"arxiv","version":1},"verdict":{"id":"4dd133c4-a98f-42ef-bbbb-7f8429e6c9d0","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T01:45:53.798729Z","strongest_claim":"We propose a general framework for obtaining planning-level privacy, which works by adding mock agents to the planning process. We show how to adapt two popular MAPF algorithms, namely PIBT and LaCAM, such that they preserve execution-level privacy. Lastly, we propose a post-processing technique that allows the agents to reduce the sum of costs of the returned solution without losing any privacy.","one_line_summary":"New algorithms for multi-agent path finding preserve planning-level privacy via mock agents and execution-level privacy via modified PIBT and LaCAM solvers, plus post-processing that lowers total cost without breaking privacy.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that inserting mock agents during planning sufficiently obscures real agent locations for all participants without introducing detectable information leaks or violating collision constraints, and that the privacy-preserving modifications to PIBT and LaCAM retain the original algorithms' correctness properties.","pith_extraction_headline":"Adding mock agents during planning lets multiple agents find collision-free paths without revealing their exact planned locations to each other."},"references":{"count":32,"sample":[{"doi":"","year":2025,"title":"Shahar Bardugo, Daniel Koyfman, and Dor Atzmon. 2025. Finding All Optimal So- lutions in Multi-Agent Path Finding. InInternational Symposium on Combinatorial Search. 20–28","work_id":"b22a45da-bf3b-4d9e-989d-8f04aae02d4e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2015,"title":"Ronen I Brafman. 2015. A privacy preserving algorithm for multi-agent planning and search. In24th International Joint Conference on Artificial Intelligence, IJCAI","work_id":"a7f28d08-6f05-41d9-9d7b-34a047154899","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"International Joint Conferences on Artificial Intelligence, 1530–1536","work_id":"5a3d1d86-74a5-4f0b-ac2a-d720f84a45b5","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Stepan Dergachev and Konstantin Yakovlev. 2021. Distributed multi-agent naviga- tion based on reciprocal collision avoidance and locally confined multi-agent path finding. InIEEE International Confere","work_id":"b61fedf3-dfc0-4791-9293-f126d1fddcfb","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2008,"title":"Boi Faltings, Thomas Léauté, and Adrian Petcu. 2008. Privacy guarantees through distributed constraint satisfaction. In2008 IEEE/WIC/ACM International Confer- ence on Web Intelligence and Intelligent ","work_id":"4af9349d-ac31-4abf-ae5d-42d3054502db","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":32,"snapshot_sha256":"0a5d9b63c4495f57423ec928eed5fd113dbb41c5d1ff2783758322a5d51421ee","internal_anchors":0},"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"}