BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Learning MPC and its applications to robotic systems
DTSTART:20191108T110000
DTEND:20191108T120000
DTSTAMP:20260603T002253Z
UID:3303469ad516dbd83284889e99d89c30cfb30cec6464de68e2adb493
CATEGORIES:Conferences - Seminars
DESCRIPTION:Ugo Rosolia from the University of California at Berkeley\nAbs
 tract\nExploiting historical data in order to iteratively improve the perf
 ormance of predictive controllers has been an active theme of research in 
 the past few decades. The key idea is to use recorded state-input pairs in
  order to compute at least one of the following three components: i) a mod
 el which describes the evolution of the system\, ii) a safe set of states 
 (and an associated control policy) from which the control task can be safe
 ly executed and iii) a value function which represents the cumulative clos
 ed-loop cost from a given state of the safe set. \nIn this talk\, I will 
 first provide an overview of the theory of Learning Model Predictive Contr
 ol that I have developed during my PhD. In particular\, I will show how hi
 storical data can be used in the control design in order to guarantee safe
 ty\, exploration and performance improvement. In the second part of the ta
 lk\, I will show the effectiveness of the proposed methodology on an auton
 omous racing example and a manipulator task example.\n\nBio\nUgo Rosolia r
 eceived the B.S. and M.S. cum laude degrees in mechanical engineering from
  the Politecnico di Milano in 2012 and 2014\, respectively. He is currentl
 y pursuing the Ph.D. degree in mechanical engineering at the University of
  California at Berkeley. \nHe was a Visiting Scholar at the Tongji Univer
 sity in Shanghai for the Double Degree Program PoliTong (Fall 2010 - Sprin
 g 2011) and at the University of Illinois at Urbana-Champaign (Fall 2013 -
  Spring 2014)\, sponsored by a Global E3 Scholarship. He was a Research En
 gineer with Siemens PLM Software in Belgium (Spring and Summer 2015). His 
 current research interests include approximate dynamic programming\, syste
 m identification\, optimization\, iterative learning control and predictiv
 e control.
LOCATION:ME C2 405 https://plan.epfl.ch/?room==ME%20C2%20405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR