-Teams link : https://teams.microsoft.com/l/meetup-join/19%3ameeting_M2I2NzE3MmYtNDYwOS00YzU3LTg4NzEtNDI0ODNjODlmMTkz%40thread.v2/0?context=%7b%22Tid%22%3a%22158716cf-46b9-48ca-8c49-c7bb67e575f3%22%2c%22Oid%22%3a%22955ae724-3c70-4f81-8a92-327e1dc5af57%22%7d
– Titre : Learning and optimization for robotic search and mapping
– Résumé : Machine learning has made great strides in the past decades, and its impact in control is increasingly being felt. In the first part of this talk, we will briefly introduce reinforcement learning, followed by customized algorithms for two problems in active robotic search and collection. The second part will adapt global optimization algorithms to the setting in which one or several robots must find the maximum of the function defined over their physical operating area, a version of source seeking. This second set of algorithms will be associated with guarantees on near-optimality as a function of the number of steps taken by the robots.
– Biographie : Lucian Busoniu received his Ph.D. degree cum laude from the Delft University of Technology, the Netherlands, in 2009. He is a full professor with the Department of Automation at the Technical University of Cluj-Napoca, where he leads the group on Robotics and Nonlinear Control. He has previously held research positions in the Netherlands and in France. His research interests include nonlinear optimal control using artificial intelligence and reinforcement learning techniques, robotics, and multiagent systems. His publications include among others a book on reinforcement learning and several influential review articles, two of which were in the top 1% cited papers in ISI Web of Knowledge. He serves on the editorial board of the Elsevier journal Engineering Applications of Artificial Intelligence, and was the recipient of the 2009 Andrew P. Sage Award for the Best Paper in the IEEE Transactions on Systems, Man, and Cybernetics. He has competitively obtained research funding in excess of 1.5 million EUR as principal investigator, and — in addition to conference keynotes — was invited to present research on prime-time national TV and in other media.
– Salle CRAN-ENSEM