{"667377":{"#nid":"667377","#data":{"type":"event","title":"MS Defense by Joshua Kuperman","body":[{"value":"\u003Cp\u003EJoshua Kuperman\u003Cbr \/\u003E\r\nAdvisor: Prof. Evangelos Theodorou will defend a master\u2019s thesis entitled,\u003Cbr \/\u003E\r\nImproved Exploration for Data-Driven, Safety-Embedded Differential Dynamic Programming Using\u0026nbsp;\u003Cbr \/\u003E\r\nTolerant Barrier States and Pontryagin Differentiable Programming\u003Cbr \/\u003E\r\nOn\u003Cbr \/\u003E\r\nThursday, April 20 at 10:00 a.m.\u003Cbr \/\u003E\r\nEngineering, Science, and Mechanics Building G08\u003Cbr \/\u003E\r\nAbstract\u003Cbr \/\u003E\r\nA great challenge exists at the intersection of perception and controls \u2013 integrating the\u0026nbsp;\u003Cbr \/\u003E\r\nuncertainty present in perception-based state and obstacle estimation into safe control and\u0026nbsp;\u003Cbr \/\u003E\r\ntrajectory optimization. First, we present the tolerant discrete barrier state (T-DBaS), a novel\u0026nbsp;\u003Cbr \/\u003E\r\nsafety-embedding technique for trajectory optimization with enhanced exploratory capabilities. This\u0026nbsp;\u003Cbr \/\u003E\r\napproach generalizes the standard discrete barrier state (DBaS) method by accommodating temporary\u0026nbsp;\u003Cbr \/\u003E\r\nconstraint violation during the optimization process while still approximating its safety\u0026nbsp;\u003Cbr \/\u003E\r\nguarantees. Towards applying T-DBaS to safety- critical autonomous robotics, we combine it with\u0026nbsp;\u003Cbr \/\u003E\r\nDifferential Dynamic Programming (DDP), leading to the proposed safe trajectory optimization method\u0026nbsp;\u003Cbr \/\u003E\r\nT-DBaS-DDP, which inherits the convergence and scalability properties of the solver. Despite this,\u0026nbsp;\u003Cbr \/\u003E\r\nthe tolerant barrier function parameters require tuning to reach peak performance for a wide array\u0026nbsp;\u003Cbr \/\u003E\r\nof constraints. To alleviate this requirement, we tune the T- DBaS parameters with the\u0026nbsp;\u003Cbr \/\u003E\r\nparameterized trajectory optimizer Pontryagin Differentiable Programming (PDP), proposing\u0026nbsp;\u003Cbr \/\u003E\r\nT-DBaS-PDP, an interpretable and generalizable solver for a variety of optimal control problems. In\u0026nbsp;\u003Cbr \/\u003E\r\norder to integrate perception uncertainty into safe optimal control, we learn the safety of the\u0026nbsp;\u003Cbr \/\u003E\r\nsystem via gaussian processes to create an interpretable, data-driven, and safety-guaranteeable\u0026nbsp;\u003Cbr \/\u003E\r\nframework. We implement this framework on differential drive and quadrotor dynamics and show its\u0026nbsp;\u003Cbr \/\u003E\r\nimprovement over hand-tuned T-DBaS-DDP.\u003Cbr \/\u003E\r\nCommittee\u003Cbr \/\u003E\r\n\u2022 \u0026nbsp;Prof. Evangelos Theodorou \u2013 School of Aerospace Engineering (advisor)\u003Cbr \/\u003E\r\n\u2022 \u0026nbsp;Prof. Kyriakos G. Vamvoudakis \u2013 School of Aerospace Engineering\u003Cbr \/\u003E\r\n\u2022 \u0026nbsp;Prof. Patricio A. Vela \u2013 School of Electrical and Computer Engineering\u003Cbr \/\u003E\r\n\u0026nbsp;\u003C\/p\u003E\r\n","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003ESee abstract\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"Improved Exploration for Data-Driven, Safety-Embedded Differential Dynamic Programming Using Tolerant Barrier States and Pontryagin Differentiable Programming"}],"uid":"27707","created_gmt":"2023-04-17 15:44:41","changed_gmt":"2023-04-17 15:44:41","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2023-04-20T10:00:00-04:00","event_time_end":"2023-04-20T12:00:00-04:00","event_time_end_last":"2023-04-20T12:00:00-04:00","gmt_time_start":"2023-04-20 14:00:00","gmt_time_end":"2023-04-20 16:00:00","gmt_time_end_last":"2023-04-20 16:00:00","rrule":null,"timezone":"America\/New_York"},"location":"Engineering, Science, and Mechanics Building G08","extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"111531","name":"ms defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"78771","name":"Public"},{"id":"174045","name":"Graduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}