BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Solving hard problems in robotics – with a little help from semi
 definite relaxations\, nullspaces\, and sparsity
DTSTART:20240312T161500
DTSTAMP:20260509T055428Z
UID:5375ad9dd1f54f791d773a7074142d29178d67ac0e94f2ac9c0ee0ce
CATEGORIES:Conferences - Seminars
DESCRIPTION:Frederike Duembgen (Université de Toronto)\nAbstract: Many st
 ate estimation and planning tasks in robotics are formulated as non-convex
  optimization problems\, and commonly deployed efficient solvers may conve
 rge to poor local minima. Recent years have seen promising developments in
  so-called certifiably optimal estimation\, showing that many problems can
  in fact be solved to global optimality or certified through the use of ti
 ght semidefinite relaxations. \nIn this talk\, I present our efforts to m
 ake such methods – for the field of state estimation in particular – m
 ore practical for roboticists. Among those efforts\, I will present novel 
 efficient optimality certificates as a low-cost add-on to off-the-shelf lo
 cal solvers\, which apply to a variety of problems including range-only\, 
 stereo-camera and\, more generally\, matrix-weighted localization. Then\, 
 I present our approach to automatically certify almost any state estimatio
 n problem\, using a sampling-based method to automatically find tight rela
 xations through nullspace characterizations. I end with an overview of our
  most recent work\, which allows to create both fast and certifiably optim
 al solvers by exploiting the sparse problem structure.\n 
LOCATION:CM1517
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR