Emo Todorov: Synthesis of contact-rich behaviors with optimal control
Brainy Days in Jerusalem:
An interdisciplinary celebration
June 22-25, 2015, Mishkenot Sha’ananim, Jerusalem, Israel
Animals and machines interact with their environment mainly through physical contact. Yet the discontinuous nature of contact dynamics complicates planning and control, especially when combined with uncertainty. We have recently made progress in terms of optimizing complex trajectories that involve many contact events. These events do not need to be specified in advance, but instead are discovered fully automatically. Key to our success is the development of new models of contact dynamics, which enable continuation methods that in turn help the optimizer avoid a combinatorial search over contact configurations. We can presently synthesize humanoid trajectories in tasks such as getting up from the floor, walking and running, turning, riding a unicycle, as well as a variety of dexterous hand manipulation tasks. When augmented with warm-starts in the context of model predictive control, our optimizers can run in real-time and be used as approximately-optimal feedback controllers. Some of these controllers have already been transferred to physical robots, via ensemble optimization methods that increase robustness to modeling errors. The resulting trajectory libraries are also used to train recurrent neural networks. After training the networks can control the body autonomously, without further help from the trajectory optimizer.
Видео Emo Todorov: Synthesis of contact-rich behaviors with optimal control канала ELSC Video
An interdisciplinary celebration
June 22-25, 2015, Mishkenot Sha’ananim, Jerusalem, Israel
Animals and machines interact with their environment mainly through physical contact. Yet the discontinuous nature of contact dynamics complicates planning and control, especially when combined with uncertainty. We have recently made progress in terms of optimizing complex trajectories that involve many contact events. These events do not need to be specified in advance, but instead are discovered fully automatically. Key to our success is the development of new models of contact dynamics, which enable continuation methods that in turn help the optimizer avoid a combinatorial search over contact configurations. We can presently synthesize humanoid trajectories in tasks such as getting up from the floor, walking and running, turning, riding a unicycle, as well as a variety of dexterous hand manipulation tasks. When augmented with warm-starts in the context of model predictive control, our optimizers can run in real-time and be used as approximately-optimal feedback controllers. Some of these controllers have already been transferred to physical robots, via ensemble optimization methods that increase robustness to modeling errors. The resulting trajectory libraries are also used to train recurrent neural networks. After training the networks can control the body autonomously, without further help from the trajectory optimizer.
Видео Emo Todorov: Synthesis of contact-rich behaviors with optimal control канала ELSC Video
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Cinema and Brain: Prof. Idan SegevLarry Abbott - Mean-field theory of networks with plastic synapsesקולנוע ומוח 2012: לגרות את הנפשFace-Selective Neurons in the Vicinity of the Human Fusiform Face AreaSight Through Sound: teaching blind people to see through soundNervous club - Cognitive and electrophysiological manifestations of implicit temporal expectationsELSC-ICNC Seminar: Sagiv ShifmanShie Mannor: Regularization and Robustness in Reinforcement LearningELSC Seminar: Dora AngelakiTomorrow: State of Mind - Adapt, Learn, and then WalkELSC Seminar Series 2020/2021: Prof. Timothy Behrens, University of Oxford, Nov. 19th, 2020Idan Segev - The MYSTERY of the Creative Brain - 05-2017היום שאחרי: מפת דרכים - פרופ' ברק מדינהDisability or Disease: Connectivity Pattern Changes in the Visual Network After Optic Nerve DamageKenji Doya: Neural Circuit Mechanisms of Mental SimulationIlluminating disorientation in Alzheimer’s disease through the use of multimodal neuroimagingInto the future: Human Brain Project Summit 2015WMUH 2018 Discussion Tishby, Shashua, Ullman and Noah HarariELSC Seminar: Ifat LevyFrontiers for young minds: Boran Salah 11 years old interviews Idan SegevBrain & Art week 2019 - Prof. Masha Niv and Chef Avner Niv - Facts, myths and tricks about taste