www.AV 电影天堂.com



2020-01-09 11:01   自动化学院 審核人:   (點擊: )

報告題目:Learning-based Optimal Control with Guaranteed Stability

報告人:姜钟平教授 (纽约大学)




报告简介:This talk considers learning-based control design for continuous-time dynamical systems with unknown dynamics. It introduces a new design paradigm, called “Robust Adaptive Dynamic Programming (RADP)”, that is fundamentally different from traditional control theory. In the classical paradigm, controllers are often designed for a given class of dynamical control systems; it is a model-based design. In the RADP paradigm, controllers are learned online using real-time input-output data collected along the trajectories of the control system in question. An entanglement of techniques from reinforcement learning and model-based control theory is advocated to find a sequence of suboptimal controllers that will approximate the optimal solution as learning steps increase. On the one hand, this RADP paradigm overcomes the well-known “curse of dimensionality” and the “curse of modeling” associated with Bellman's Dynamic Programming. On the other hand, rigorous stability and robustness analysis can be derived for the closed-loop system with real-time learning-based controllers. The effectiveness of RADP as a new framework for data-driven nonlinear control design is demonstrated via its applications to power systems, autonomous vehicles, and biological motor control.


姜钟平,自1999年受聘于美国纽约科技大学(现合并成为纽约大学工学院)以来,历任助理教授、副教授和教授。是非线性小增益理论的关键创立者,主要研究领域为稳定性理论、鲁棒自适应非线性控制及其在通讯网络、欠驱动力学系统、多智能体和系统神经学中的应用。基于他的研究工作,姜博士于1998年获得澳大利亚伊丽莎白二世杰出研究奖,2001年获得美国国家科学基金会成就奖,2005年获日本科学振兴会研究奖, 2007年获得中国自然科学基金会海外杰出华人研究奖(中科院系统科學研究所)。2008年当选为美国电气电子工程师协会会士。2009年获选教育部“长江学者”讲座教授(北京大学)。2013年当选为国际自动控制联合会会士。2017年当选为中国自动化学会会士。2018年入选Clarivate Analytics高被引研究员。