BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Invariant Kalman filtering - Silvère Bonnabel\, Mines ParisTech
DTSTART:20171116T140000Z
DTEND:20171116T150000Z
UID:TALK81521@talks.cam.ac.uk
CONTACT:Tim Hughes
DESCRIPTION:The Kalman filter\, or more precisely the extended Kalman filt
 er (EKF)\, is a fundamental engineering tool that is pervasively used in c
 ontrol\, robotics\, and for various estimation tasks in autonomous systems
 . The recent field of Invariant extended Kalman filtering\, aims at using 
 the geometric structure of the state space and the dynamics to improve the
  EKF\, notably in terms of mathematical guarantees. The methodology essent
 ially applies in the field of localization\, navigation\, and simultaneous
  localization and mapping (SLAM) where it is proved to resolve the well-kn
 own inconsistencies of the conventional EKF. Albeit recent\, its remarkabl
 e robustness properties have already prompted a true industrial implementa
 tion in the aerospace field. \n\nThis talk aims to provide an intuitive in
 troduction to the methodology of invariant Kalman filtering\, to underline
  what the important differences with the conventional EKF are\, and to giv
 e the main reasons why it resolves the EKF consistency issues for SLAM. Th
 is should be of interest to students or researchers intrigued by the appli
 cation of  mathematical theories to practical applications\, or interested
  in finding simple to implement and robust filters for localization\, navi
 gation\, and SLAM\, notably for autonomous vehicle guidance.
LOCATION:Cambridge University Engineering Department\, Lecture Room number
  LR11
END:VEVENT
END:VCALENDAR