BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Sensor fusion and parameter inference in nonlinear dynamical syste
 ms - Thomas B. Schön\,   Associate Professor\, Linköping University
DTSTART:20130418T100000Z
DTEND:20130418T113000Z
UID:TALK44505@talks.cam.ac.uk
CONTACT:Dr Jason Z JIANG
DESCRIPTION:In this talk I will provide an overview of some of the work we
  do when it comes to solving inference\nproblems in nonlinear dynamical sy
 stems. A proper subtitle for the talk is "strategies and\nexamples"\, sinc
 e I will only provide solution strategies and show that these strategies c
 an\nsuccessfully solve challenging applications. The first topic is parame
 ter inference problems in\nnonlinear dynamical systems (a.k.a. nonlinear s
 ystem identification). The maximum likelihood problem\nis solved using a c
 ombination of the expectation maximization (EM) algorithm and sequential M
 onte\nCarlo (SMC) methods (e.g.\, the particle filter and the particle smo
 other). The Bayesian problem is\nsolved using a combination of Markov chai
 n Monte Carlo (MCMC) and SMC. As an example\, we show how to\nestimate a p
 articular special case known as the Wiener model (a linear dynamical model
  followed by a\nstatic nonlinearity). The second topic is that of sensor f
 usion\, which refers to the problem of\ninferring states (and possibly par
 ameters) using measurements from several different\, often\ncomplementary\
 , sensors. The strategy is explained and (perhaps more importantly) illust
 rated using\nthree of the industrial applications we are working with\; 1.
  Navigation of fighter aircraft (using\ninertial sensors\, radar and maps)
 \; 2. Indoor positioning of humans (using inertial sensors and\nmaps)\; 3.
  Indoor pose estimation of a human body (using inertial sensors and ultra-
 wideband).
LOCATION:Cambridge University Engineering Department\, LR3A
END:VEVENT
END:VCALENDAR