{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/multimodal-probabilistic-model-based-planning","title":"Multimodal Probabilistic Model-Based Planning for Human-Robot Interaction","arxiv_id":"1710.09483","date":"2017-10-25","proceeding":null,"authors":["Edward Schmerling","Karen Leung","Wolf Vollprecht","Marco Pavone"],"abstract":"This paper presents a method for constructing human-robot interaction\npolicies in settings where multimodality, i.e., the possibility of multiple\nhighly distinct futures, plays a critical role in decision making. We are\nmotivated in this work by the example of traffic weaving, e.g., at highway\non-ramps/off-ramps, where entering and exiting cars must swap lanes in a short\ndistance---a challenging negotiation even for experienced drivers due to the\ninherent multimodal uncertainty of who will pass whom. Our approach is to learn\nmultimodal probability distributions over future human actions from a dataset\nof human-human exemplars and perform real-time robot policy construction in the\nresulting environment model through massively parallel sampling of human\nresponses to candidate robot action sequences. Direct learning of these\ndistributions is made possible by recent advances in the theory of conditional\nvariational autoencoders (CVAEs), whereby we learn action distributions\nsimultaneously conditioned on the present interaction history, as well as\ncandidate future robot actions in order to take into account response dynamics.\nWe demonstrate the efficacy of this approach with a human-in-the-loop\nsimulation of a traffic weaving scenario.","url_abs":"http://arxiv.org/abs/1710.09483v1","url_pdf":"http://arxiv.org/pdf/1710.09483v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"multimodal-probabilistic-model-based-planning","repo_url":"https://github.com/StanfordASL/TrafficWeavingCVAE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1710.09483","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}