University of Cambridge > Talks.cam > Signal Processing and Communications Lab Seminars > Dynamic State Estimation using Dirac Mixture Approximation and Directional Statistics

Dynamic State Estimation using Dirac Mixture Approximation and Directional Statistics

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Prof. Ramji Venkataramanan.

In many applications of stochastic filtering, it is of interest to consider inherently nonlinear system models or inherently nonlinear domains such as the sphere or the circle. This motivates the development of nonlinear estimation techniques that are able to capture the nonlinearity and at the same time are based on a valid distributional assumption.

Propagation of continuous probability distributions through nonlinear functions might be numerically burdensome and not solvable in closed form. Thus, we propose a general framework for approximating a given probability distribution by another distribution. This is also of interest in other areas such as model predictive control or information theory. Furthermore, we discuss scenarios where the Gaussian assumption might be inherently invalid which might happen for estimation of angles or orientation involving highly uncertain measurements or strong system noise. Thus, filters based on circular and spherical distributions are proposed in order to handle this kind of problems. Finally, we will discuss the challenges involved in combining non-Gaussian distributional assumptions and approximate uncertainty propagation techniques.

BIO: Igor Gilitschenski received his diploma in mathematics (major) and computer science (minor) from the University of Stuttgart in September 2011. In November 2011, he joined the Intelligent Sensor-Actuator Systems (ISAS) Laboratory at the Karlsruhe Institute of Technology (KIT), where he is working on his PhD within the research training group “Self-organizing Sensor- Actuator Networks”. Igor co-authored a paper, which received the “Best Student Paper Award, First Runner-Up” at the 16th International Conference on Information Fusion.

This talk is part of the Signal Processing and Communications Lab Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity