Laurent Lessard, Nov 2013

From CDSWiki
Jump to navigationJump to search

Laurent Lessard, a postdoctoral researcher in Mechanical Engineering at UC Berkeley, working with Andrew Packard and Kameshwar Poolla, will be on campus Thursday, November 21st and Friday, November 22nd. His research interests include decentralized control, optimization, and finding computationally tractable approaches for complex engineering applications.

Please add yourself and a location to the schedule below, if you would like to meet with him. He will have an office in 202 Annenberg, during his visit.


Thursday, November 21st

  • 10:00 AM
  • 10:45 AM
  • 12:00 PM - Lunch (join John Doyle at the Murray group meeting in 213 Annenberg)
  • 1:15 PM - Nikolai Matni, 335 Annenberg
  • 2:00 PM - Give Seminar, 314 Annenberg
  • 3:30 PM - Steven Low, 219 Annenberg
  • 4:00 PM - Attend seminar in Beckman Lab 24
  • 5:00 PM - Seungil You ANB 231
  • 5:45 PM - Done
  • 6:30 PM - Dinner with John Doyle's students

Friday, November 22nd

  • 10:00 AM - Daniela Meola, 223 Annenberg
  • 10:45 AM - Nikolai Matni, 335 Annenberg
  • 12:00 PM - Lunch with John Doyle
  • 1:15 PM
  • 2:00 PM - Matanya Horowitz
  • 2:45 PM - Eric Wolff
  • 3:15 PM
  • 4:00 PM - Depart campus from transportation lot

Talk Abstract

H2 and Hinfinity optimal two-player output feedback

In this talk, we consider a fundamental decentralized optimal control problem: two interconnected linear subsystems with a partially nested information pattern and output feedback. Our main contribution is an explicit generically-minimal state-space realization of the H2-optimal controller, which was previously not known. The solution we present provides much more than just a formula; it gives us the state dimension of the optimal controller, and reveals precisely what sort of estimation and control structure is optimal. This structure is new and different from the classical LQG result, yet there is substantial increase in the computational complexity of finding the optimal controller. We also give a brief preview of how these results extend to the suboptimal-Hinfinity case.