{"592395":{"#nid":"592395","#data":{"type":"event","title":"Ph.D. Thesis Proposal: Daehyung Park","body":[{"value":"\u003Cp\u003EAssistive robots have the potential to serve as\u0026nbsp;caregivers, assisting with activities of daily living (ADLs) and\u0026nbsp;instrumental activities of daily living (IADLs). Detecting when\u0026nbsp;something has gone wrong could help assistive robots operate more\u0026nbsp;safely and effectively around people. However, the complexity of\u0026nbsp;interacting with people and objects in human environments can\u0026nbsp;make errors difficult to detect. We introduce a multimodal\u0026nbsp;execution monitoring system to detect and classify anomalous\u0026nbsp;executions when robots operate near humans. The system\u0026rsquo;s\u0026nbsp;anomaly detector models multimodal sensory signals with a hidden\u0026nbsp;Markov model (HMM) and uses a likelihood threshold that varies based on the progress of task execution. The system classifies\u0026nbsp;the type and cause of common anomalies using an artificial neural\u0026nbsp;network. We evaluate my system with haptic, visual, auditory, and\u0026nbsp;kinematic sensing during household tasks and human-robot interactive tasks (feeding assistance) performed by a PR2 robot\u0026nbsp;with able-bodied participants and people with disabilities. In\u0026nbsp;our evaluation, our methods performed better than other methods\u0026nbsp;from the literature, yielding higher area under curve (AUC) and\u0026nbsp;shorter detection delays. Multimodality also improved the\u0026nbsp;performance of monitoring methods by detecting a broader range of\u0026nbsp;anomalies.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EPh.D. Thesis Proposal (Daehyung Park)\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"Ph.D. Thesis Proposal (Daehyung Park)"}],"uid":"27469","created_gmt":"2017-06-02 18:24:08","changed_gmt":"2017-06-02 18:24:23","author":"Kristen Bailey","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2017-06-07T12:00:00-04:00","event_time_end":"2017-06-07T14:00:00-04:00","event_time_end_last":"2017-06-07T14:00:00-04:00","gmt_time_start":"2017-06-07 16:00:00","gmt_time_end":"2017-06-07 18:00:00","gmt_time_end_last":"2017-06-07 18:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"1182","name":"General"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"78771","name":"Public"},{"id":"174045","name":"Graduate students"},{"id":"78751","name":"Undergraduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EDaehyung Park\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Ca href=\u0022mailto:deric.park@gatech.edu\u0022\u003Ederic.park@gatech.edu\u003C\/a\u003E\u003C\/p\u003E\r\n","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}