{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,22]],"date-time":"2026-01-22T08:49:09Z","timestamp":1769071749783,"version":"3.49.0"},"reference-count":73,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2025,9,18]],"date-time":"2025-09-18T00:00:00Z","timestamp":1758153600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,9,18]],"date-time":"2025-09-18T00:00:00Z","timestamp":1758153600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["advanced.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Advanced Intelligent Systems"],"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:p>A key goal in evacuation management is to quickly and safely remove panicking crowds from buildings, festivals, or airplanes while preventing crush fatalities. Recently, there has been much progress in realistically modeling crowds in complex environments, based on social force models, cellular automata, and machine learning. However, current models assume specific social interactions and do not allow to systematically explore how to optimize crowd cooperation and evacuation. In contrast, the present work focuses on the question, how an ideal crowd of superintelligent agents, comprising humans, robots, or smart active particles, would cooperate to optimize evacuation. A method is developed that uses multiagent reinforcement learning combined with self\u2010play to learn optimal crowd behavior from scratch. Crucially, the agents in this approach are pressure\u2010aware and autonomously learn collision and crushing avoidance. After training, they adopt interpretable evacuation strategies like queuing and zipper merging and outperform traditional evacuation models in terms of fatality avoidance and evacuation rate. Our method can be used to enhance guidelines for mass evacuation, potentially saving lives.<\/jats:p>","DOI":"10.1002\/aisy.202500436","type":"journal-article","created":{"date-parts":[[2025,9,18]],"date-time":"2025-09-18T17:46:01Z","timestamp":1758217561000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Learning Optimal Crowd Evacuation from Scratch Through Self\u2010Play"],"prefix":"10.1002","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6538-754X","authenticated-orcid":false,"given":"Mahdi","family":"Nasiri","sequence":"first","affiliation":[{"name":"Institute of Condensed Matter Physics Technische Universit\u00e4t Darmstadt  64289 Darmstadt Germany"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-0568-0275","authenticated-orcid":false,"given":"Malte","family":"Cordts","sequence":"additional","affiliation":[{"name":"Institute of Condensed Matter Physics Technische Universit\u00e4t Darmstadt  64289 Darmstadt Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8305-9379","authenticated-orcid":false,"given":"Heinz","family":"Koeppl","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Information Technology Technische Universit\u00e4t Darmstadt  64283 Darmstadt Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7647-6430","authenticated-orcid":false,"given":"Benno","family":"Liebchen","sequence":"additional","affiliation":[{"name":"Institute of Condensed Matter Physics Technische Universit\u00e4t Darmstadt  64289 Darmstadt Germany"}]}],"member":"311","published-online":{"date-parts":[[2025,9,18]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"crossref","unstructured":"C. W.Reynolds inProc. of the 14th Annual Conf. on Computer Graphics and Interactive Techniques Association for Computing Machinery New York NY1987 pp.25\u201334.","DOI":"10.1145\/37401.37406"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.physrep.2012.03.004"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.copsyc.2020.02.005"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-conmatphys-031620-100450"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/TETCI.2020.2992778"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1080\/01441647.2015.1017866"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1061\/(ASCE)IS.1943-555X.0000080"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4939-2565-0_60"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1177\/15501477211018140"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.75.046109"},{"key":"e_1_2_9_12_1","first-page":"RRN1301","volume":"4","author":"Soomaroo L.","year":"2012","journal-title":"PLoS Curr."},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijinfomgt.2018.05.004"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-022-04900-x"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1017\/S1049023X18001073"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1103\/RevModPhys.73.1067"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.trc.2013.02.005"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.fluid.35.101101.161136"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/1141911.1142008"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.51.4282"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1038\/35035023"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1080\/01441647.2017.1396265"},{"key":"e_1_2_9_23_1","first-page":"6","volume":"141","author":"Ghosh P. K.","year":"2014","journal-title":"J. Chem. Phys."},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.94.012117"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1063\/5.0070842"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.102.042617"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1039\/D4SM00045E"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.110.064102"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0378-4371(01)00141-8"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.physa.2019.03.117"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.trb.2017.06.017"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.91.022808"},{"key":"e_1_2_9_33_1","doi-asserted-by":"crossref","unstructured":"J.Van den Berg M.Lin D.Manocha in2008 IEEE Int. Conf. on Robotics and Automation IEEE Piscataway NJ2008 pp.1928\u20131935.","DOI":"10.1109\/ROBOT.2008.4543489"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1016507108"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1017\/S0263574714000289"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.89.032806"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.camwa.2011.10.048"},{"key":"e_1_2_9_38_1","unstructured":"K.Cui S.Hauck C.Fabian H.Koeppl inICLR 2024: The Twelfth Int. Conf. on Learning Representations OpenReview MA USA2024."},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.120254"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ssci.2021.105215"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.trc.2022.103699"},{"key":"e_1_2_9_42_1","doi-asserted-by":"crossref","unstructured":"A.Alahi K.Goel V.Ramanathan A.Robicquet L.Fei\u2010Fei S.Savarese inProc. of the IEEE Conf. on Computer Vision and Pattern Recognition IEEE Piscataway NJ2016 pp.961\u2013971.","DOI":"10.1109\/CVPR.2016.110"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11069-020-04124-3"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-022-07758-5"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1209\/0295-5075\/127\/64003"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.1017\/jfm.2022.476"},{"key":"e_1_2_9_47_1","volume-title":"Active Colloids: From Fundamentals to Frontiers","author":"Volpe G.","year":"2024"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.simpat.2014.06.005"},{"key":"e_1_2_9_49_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.physa.2021.125845"},{"key":"e_1_2_9_50_1","doi-asserted-by":"crossref","unstructured":"Z.Sui Z.Pu J.Yi X.Tan in2018 Int. Joint Conf. on Neural Networks (IJCNN) IEEE Piscataway NJ2018 1\u20138.","DOI":"10.1109\/IJCNN.2018.8489066"},{"key":"e_1_2_9_51_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2019.08.021"},{"key":"e_1_2_9_52_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSMC.2020.2967936"},{"key":"e_1_2_9_53_1","doi-asserted-by":"publisher","DOI":"10.1080\/17538947.2023.2182376"},{"key":"e_1_2_9_54_1","doi-asserted-by":"publisher","DOI":"10.1177\/0278364915619772"},{"key":"e_1_2_9_55_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11721-017-0142-9"},{"key":"e_1_2_9_56_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cogr.2023.07.004"},{"key":"e_1_2_9_57_1","doi-asserted-by":"publisher","DOI":"10.1287\/trsc.1040.0108"},{"key":"e_1_2_9_58_1","doi-asserted-by":"publisher","DOI":"10.1038\/srep07324"},{"key":"e_1_2_9_59_1","volume-title":"Reinforcement Learning and Optimal Control","author":"Bertsekas D.","year":"2019"},{"key":"e_1_2_9_60_1","first-page":"21","volume-title":"Simulation of Pedestrian Crowds in Normal and Evacuation Situations","author":"Helbing D.","year":"2002"},{"key":"e_1_2_9_61_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.physa.2020.125186"},{"key":"e_1_2_9_62_1","doi-asserted-by":"publisher","DOI":"10.1287\/trsc.1040.0102"},{"key":"e_1_2_9_63_1","doi-asserted-by":"publisher","DOI":"10.1287\/trsc.1090.0263"},{"key":"e_1_2_9_64_1","doi-asserted-by":"publisher","DOI":"10.1063\/1.1674820"},{"key":"e_1_2_9_65_1","unstructured":"J.Heinrich D.Silver(Preprint) arXiv:1603.01121 v1 Submitted: Mar.2016."},{"key":"e_1_2_9_66_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aar6404"},{"key":"e_1_2_9_67_1","unstructured":"C.Berner G.Brockman B.Chan V.Cheung P.Debiak C.Dennison D.Farhi Q.Fischer S.Hashme C.Hesse R.J\u00f3zefowicz S.Gray C.Olsson J.Pachocki M.Petrov H.Pinto J.Raiman T.Salimans J.Schlatter J.Schneider S.Sidor I.Sutskever J.Tang F.Wolski S.Zhang(Preprint) arXiv:1912.06680 Submitted: Dec.2019."},{"key":"e_1_2_9_68_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12544-017-0264-6"},{"key":"e_1_2_9_69_1","doi-asserted-by":"publisher","DOI":"10.1109\/TAES.2016.140952"},{"key":"e_1_2_9_70_1","unstructured":"J.Schulman F.Wolski P.Dhariwal A.Radford O.Klimov(Preprint) arXiv:1707.06347 Submitted: Jul.2017."},{"key":"e_1_2_9_71_1","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton R. S.","year":"2018"},{"key":"e_1_2_9_72_1","unstructured":"J.Schulman P.Moritz S.Levine M.Jordan P.Abbeel(Preprint) arXiv:1506.02438 Submitted: Jun.2015."},{"key":"e_1_2_9_73_1","volume-title":"Deep Learning","author":"Goodfellow I.","year":"2016"},{"key":"e_1_2_9_74_1","first-page":"8026","volume":"32","author":"Paszke A.","year":"2019","journal-title":"Adv. Neural Info. Proces. Syst."}],"container-title":["Advanced Intelligent Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/aisy.202500436","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/aisy.202500436","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/aisy.202500436","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,21]],"date-time":"2026-01-21T16:53:15Z","timestamp":1769014395000},"score":1,"resource":{"primary":{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/10.1002\/aisy.202500436"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,18]]},"references-count":73,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1002\/aisy.202500436"],"URL":"https:\/\/doi.org\/10.1002\/aisy.202500436","archive":["Portico"],"relation":{},"ISSN":["2640-4567","2640-4567"],"issn-type":[{"value":"2640-4567","type":"print"},{"value":"2640-4567","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,18]]},"assertion":[{"value":"2025-04-17","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-18","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"2500436"}}