BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Dynamic State Estimation using Dirac Mixture Approximation and Dir
 ectional  Statistics  - Igor Gilitschenski\,  Karlsruhe Institute of Techn
 ology
DTSTART:20131128T140000Z
DTEND:20131128T150000Z
UID:TALK48765@talks.cam.ac.uk
CONTACT:Prof. Ramji Venkataramanan
DESCRIPTION:In many applications of stochastic filtering\, it is of intere
 st to consider \ninherently nonlinear system models or inherently nonlinea
 r domains such as the \nsphere or the circle. This motivates the developme
 nt of nonlinear estimation  \ntechniques that are able to capture the nonl
 inearity and at the same time are \nbased on a valid distributional assump
 tion.\n\nPropagation of continuous probability distributions through nonli
 near \nfunctions might be numerically burdensome and not solvable in close
 d form. \nThus\, we propose a general framework for approximating a given 
 probability \ndistribution by another distribution. This is also of intere
 st in other areas \nsuch as model predictive control or information theory
 . Furthermore\, we \ndiscuss scenarios where the Gaussian assumption might
  be inherently invalid \nwhich might happen for estimation of angles  or o
 rientation involving highly \nuncertain measurements or strong system nois
 e. Thus\, filters based on circular \nand spherical distributions are prop
 osed in order to handle this kind of \nproblems. Finally\, we will discuss
  the challenges involved in combining non-Gaussian distributional assumpti
 ons and approximate uncertainty propagation \ntechniques.\n\n\n*BIO*:  Igo
 r Gilitschenski received his diploma in mathematics (major) and computer \
 nscience (minor) from the University of Stuttgart in September 2011. In \n
 November 2011\, he joined the Intelligent Sensor-Actuator Systems (ISAS) \
 nLaboratory at the Karlsruhe Institute of Technology (KIT)\, where he is w
 orking \non his PhD within the research training group "Self-organizing Se
 nsor-\nActuator Networks". Igor co-authored a  paper\, which received the 
 "Best Student Paper Award\, First Runner-Up" at the 16th International Con
 ference on Information Fusion.
LOCATION:BE4-38 (CBL Meeting Room)
END:VEVENT
END:VCALENDAR