{"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/learning-models-for-shared-control-of-human","title":"Learning Models for Shared Control of Human-Machine Systems with Unknown Dynamics","arxiv_id":"1808.08268","date":"2018-08-24","proceeding":null,"authors":["Alexander Broad","Todd Murphey","Brenna Argall"],"abstract":"We present a novel approach to shared control of human-machine systems. Our\nmethod assumes no a priori knowledge of the system dynamics. Instead, we learn\nboth the dynamics and information about the user's interaction from observation\nthrough the use of the Koopman operator. Using the learned model, we define an\noptimization problem to compute the optimal policy for a given task, and\ncompare the user input to the optimal input. We demonstrate the efficacy of our\napproach with a user study. We also analyze the individual nature of the\nlearned models by comparing the effectiveness of our approach when the\ndemonstration data comes from a user's own interactions, from the interactions\nof a group of users and from a domain expert. Positive results include\nstatistically significant improvements on task metrics when comparing a\nuser-only control paradigm with our shared control paradigm. Surprising results\ninclude findings that suggest that individualizing the model based on a user's\nown data does not effect the ability to learn a useful dynamic system. We\nexplore this tension as it relates to developing human-in-the-loop systems\nfurther in the discussion.","url_abs":"http://arxiv.org/abs/1808.08268v1","url_pdf":"http://arxiv.org/pdf/1808.08268v1.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":"learning-models-for-shared-control-of-human","repo_url":"https://github.com/asbroad/koopman_operator_model_learning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.08268","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}