Learning representations that generalize across tasks and domains is challenging yet necessary for autonomous systems. Although task-driven approaches are appealing, designing models specific to each application can be difficult in the face of limited data, especially when dealing with highly variable multimodal input spaces arising from different tasks in different environments. We introduce the first general-purpose pretraining pipeline, COntrastive Multimodal Pretraining for AutonomouS Systems (COMPASS), to overcome the limitations of task-specific models and existing pretraining approaches. COMPASS constructs a multimodal graph by considering the essential information for autonomous systems and the properties of different modalities. Through this graph, multimodal signals are connected and mapped into two factorized spatiotemporal latent spaces: a \u201cmotion pattern space\u201d and a \u201ccurrent state space.\u201d By learning from multimodal correspondences in each latent space, COMPASS creates state representations that models necessary information such as temporal dynamics, geometry, and semantics. We pretrain COMPASS on a largescale multimodal simulation dataset TartanAir [1] and evaluate it on drone navigation, vehicle racing, and visual odometry tasks. The experiments indicate that COMPASS can tackle all three scenarios and can also generalize to unseen environments and real-world data.<\/p>\n","protected":false},"excerpt":{"rendered":"
Learning representations that generalize across tasks and domains is challenging yet necessary for autonomous systems. Although task-driven approaches are appealing, designing models specific to each application can be difficult in the face of limited data, especially when dealing with highly variable multimodal input spaces arising from different tasks in different environments. We introduce the first […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13561,13556,13562],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-field-of-study":[246694,246658,248587,249835],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-821134","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-algorithms","msr-research-area-artificial-intelligence","msr-research-area-computer-vision","msr-locale-en_us","msr-field-of-study-artificial-intelligence","msr-field-of-study-deep-learning","msr-field-of-study-perception","msr-field-of-study-robotics"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2022-10","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"IEEE","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/02\/COMPASS_paper.pdf","id":"831604","title":"compass_paper","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[{"id":831604,"url":"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/COMPASS_paper.pdf"},{"id":831601,"url":"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/COMPASS.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"Shuang Ma","user_id":40411,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Shuang Ma"},{"type":"user_nicename","value":"Sai Vemprala","user_id":39180,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Sai Vemprala"},{"type":"text","value":"Wenshan Wang","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Jayesh Gupta","user_id":41431,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jayesh Gupta"},{"type":"user_nicename","value":"Yale Song","user_id":37422,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Yale Song"},{"type":"user_nicename","value":"Daniel McDuff","user_id":36428,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Daniel McDuff"},{"type":"user_nicename","value":"Ashish Kapoor","user_id":30903,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Ashish Kapoor"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[867219],"msr_project":[],"publication":[],"video":[],"download":[821371],"msr_publication_type":"inproceedings","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/821134"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":4,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/821134\/revisions"}],"predecessor-version":[{"id":831607,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/821134\/revisions\/831607"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=821134"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=821134"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=821134"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=821134"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=821134"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=821134"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=821134"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=821134"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=821134"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=821134"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=821134"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=821134"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=821134"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=821134"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=821134"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}