{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T09:33:33Z","timestamp":1763458413962,"version":"3.41.2"},"reference-count":41,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2025,1,7]],"date-time":"2025-01-07T00:00:00Z","timestamp":1736208000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Journal of Field Robotics"],"published-print":{"date-parts":[[2025,8]]},"abstract":"<jats:title>ABSTRACT<\/jats:title><jats:p>Real\u2010world robots will face a wide variety of complex environments when performing navigation or exploration tasks, especially in situations where the robots have never been seen before. Usually, robots need to establish local or global maps and then use path planning algorithms to determine their routes. However, in some environments, such as a wild grassy path or pavement on either side of a road, it is difficult for robots to plan routes through navigation maps. To address this, we propose a robust framework for robot navigation using contrastive learning called Contrastive Observation\u2013Action in Latent (COAL) space. To extract features from the action space and observation space, respectively, COAL uses two different encoders. At the training stage, COAL does not require any data annotation and a mask approach is employed to keep features with significant differences away from each other in latent space. Similar to multimodal contrastive learning, we maximize bidirectional mutual information to align the features of observations and action sequences in latent space, which can enhance the generalization of the model. At the deployment stage, robots only need the current image as observation to complete exploration tasks. The most suitable action sequence is selected from the sampled data for generating control signals. We evaluate the robustness of COAL in both simulation and real environments. Only 41\u2009min of unlabeled training data is required to allow COAL to explore environments that have never been seen before, even at night. Compared with state\u2010of\u2010the\u2010art methods, COAL has the strongest robustness and generalization ability. More importantly, the robustness of COAL is further improved by augmenting our training data using other open\u2010source data sets, which indicates that our framework has great potential to extract deep features of observations and action sequences. Our code and trained models are available at <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/wzm206\/COAL\">https:\/\/github.com\/wzm206\/COAL<\/jats:ext-link>.<\/jats:p>","DOI":"10.1002\/rob.22508","type":"journal-article","created":{"date-parts":[[2025,1,7]],"date-time":"2025-01-07T06:41:32Z","timestamp":1736232092000},"page":"2028-2041","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["COAL: Robust Contrastive Learning\u2010Based Visual Navigation Framework"],"prefix":"10.1002","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-7029-6546","authenticated-orcid":false,"given":"Zengmao","family":"Wang","sequence":"first","affiliation":[{"name":"School of Advanced Interdisciplinary Sciences University of Chinese Academy of Sciences Beijing China"},{"name":"Institute of Automation Chinese Academy of Sciences Beijing China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianhua","family":"Hu","sequence":"additional","affiliation":[{"name":"Institute of Automation Chinese Academy of Sciences Beijing China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qifei","family":"Tang","sequence":"additional","affiliation":[{"name":"Institute of Automation Chinese Academy of Sciences Beijing China"},{"name":"School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2257-5684","authenticated-orcid":false,"given":"Wei","family":"Gao","sequence":"additional","affiliation":[{"name":"Institute of Automation Chinese Academy of Sciences Beijing China"},{"name":"School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2025,1,7]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/IRDS.2002.1041446"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00396"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1177\/0278364915614638"},{"key":"e_1_2_9_5_1","first-page":"1597","volume-title":"International Conference on Machine Learning, Vienna, Austria","author":"Chen T.","year":"2020"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.026"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2017.2651163"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2018.8593846"},{"key":"e_1_2_9_9_1","first-page":"1","volume-title":"Conference on Robot Learning, Mountain View, California, USA","author":"Dosovitskiy A.","year":"2017"},{"volume-title":"Proceedings of the 38th International Conference on Machine Learning (ICML\u201021)","year":"2021","author":"Du Y.","key":"e_1_2_9_10_1"},{"key":"e_1_2_9_11_1","first-page":"158","volume-title":"Conference on Robot Learning, London, United Kingdom","author":"Florence P.","year":"2022"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00975"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"volume-title":"International Conference on Learning Representations, New Orleans, LA, USA","year":"2019","author":"Hjelm R. D.","key":"e_1_2_9_14_1"},{"key":"e_1_2_9_15_1","unstructured":"Hu Y. K.Li P.Liang et al.2023. \u201cImitation With Spatial\u2013Temporal Heatmap: 2nd Place Solution for Nuplan Challenge.\u201darXiv preprint arXiv:2306.15700.https:\/\/api.semanticscholar.org\/CorpusID:259274795."},{"key":"e_1_2_9_16_1","first-page":"17853","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada","author":"Hu Y.","year":"2023"},{"volume-title":"International Conference on Machine Learning, Baltimore, Maryland, USA","year":"2022","author":"Janner M.","key":"e_1_2_9_17_1"},{"key":"e_1_2_9_18_1","first-page":"2130","volume-title":"IEEE International Conference on Robotics and Automation (ICRA)","author":"Jatavallabhula K. M.","year":"2019"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2016"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3060404"},{"key":"e_1_2_9_21_1","article-title":"A Tutorial on Energy\u2010Based Learning","volume":"1","author":"LeCun Y.","year":"2006","journal-title":"Predicting Structured Data"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1098\/rstb.2021.0447"},{"key":"e_1_2_9_23_1","first-page":"12888","volume-title":"International Conference on Machine Learning","author":"Li J.","year":"2022"},{"key":"e_1_2_9_24_1","first-page":"1","volume-title":"European Conference on Computer Vision","author":"Li Z.","year":"2022"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2024.3359548"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160968"},{"volume-title":"International Conference on Learning Representations, New Orleans, LA, USA","year":"2019","author":"Loshchilov I.","key":"e_1_2_9_27_1"},{"key":"e_1_2_9_28_1","unstructured":"Miao R. W.Liu M.Chen et al.2023. \u201cOccDepth: A Depth\u2010Aware Method for 3D Semantic Scene Completion.\u201darXiv preprint arXiv:2302.13540.https:\/\/api.semanticscholar.org\/CorpusID:257219825."},{"key":"e_1_2_9_29_1","first-page":"5171","volume-title":"International Conference on Machine Learning, Long Beach, California, USA","author":"Poole B.","year":"2019"},{"key":"e_1_2_9_30_1","first-page":"8748","volume-title":"International Conference on Machine Learning","author":"Radford A.","year":"2021"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"volume-title":"5th Annual Conference on Robot Learning, London, UK","year":"2021","author":"Shah D.","key":"e_1_2_9_32_1"},{"volume-title":"Proceedings of Robotics: Science and Systems, New York City, NY, USA","year":"2022","author":"Shah D.","key":"e_1_2_9_33_1"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10161227"},{"volume-title":"7th Annual Conference on Robot Learning, Atlanta, USA. PMLR","year":"2023","author":"Shah D.","key":"e_1_2_9_35_1"},{"key":"e_1_2_9_36_1","first-page":"63","volume-title":"2024 IEEE International Conference on Robotics and Automation (ICRA)","author":"Sridhar A.","year":"2023"},{"key":"e_1_2_9_37_1","unstructured":"Ta D.\u2010N. E.Cousineau H.Zhao andS.Feng.2022. \u201cConditional Energy\u2010Based Models for Implicit Policies: The Gap Between Theory and Practice.\u201darXiv preprint arXiv:2207.05824.https:\/\/api.semanticscholar.org\/CorpusID:250491929."},{"volume-title":"Robotics: Science and Systems, Freiburg im Breisgau","year":"2019","author":"Tabib W.","key":"e_1_2_9_38_1"},{"key":"e_1_2_9_39_1","first-page":"6105","volume-title":"International Conference on Machine Learning, Long Beach, California, USA","author":"Tan M.","year":"2019"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58621-8_45"},{"volume-title":"International Conference on Learning Representations, Addis Ababa, Ethiopia","year":"2020","author":"Tschannen M.","key":"e_1_2_9_41_1"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2020.XVI.035"}],"container-title":["Journal of Field Robotics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/rob.22508","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T06:46:51Z","timestamp":1753253211000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/rob.22508"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,7]]},"references-count":41,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,8]]}},"alternative-id":["10.1002\/rob.22508"],"URL":"https:\/\/doi.org\/10.1002\/rob.22508","archive":["Portico"],"relation":{},"ISSN":["1556-4959","1556-4967"],"issn-type":[{"type":"print","value":"1556-4959"},{"type":"electronic","value":"1556-4967"}],"subject":[],"published":{"date-parts":[[2025,1,7]]},"assertion":[{"value":"2024-06-28","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-12-14","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-01-07","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}