Tohid Ardeshiri

Since January 2016 I have been a postdoctoral fellow at the Division of Automatic Control at the Department of Electrical Engineering at Linköpings universitet (Linköping University).

Background

Between June 2009 and December 2015, I have been a PhD student at the Division of Automatic Control at the Department of Electrical Engineering at Linköpings universitet (Linköping University). I received my Master's degree (Civilingenjör) in Automotive Engineering from Chalmers University of Technology in 2005. I did my Master's thesis at Autoliv Research on an active safety application. Between February 2006 and June 2009 I was working as a senior project engineer at Autoliv Sweden AB on product development in the area of mechatronics and specifically piezoelectric technology.

I have been on paternity leave for about 6 month from my position as a PhD student.

I got my Bachelor's degree from Sharif University of Technology in Mechanical Engineering in 2003.

Research

My research interests are in analytical approximations for Bayesian learning. I am interested in various aspects of Bayesian learning, such as variational Bayes, maximum entropy kernels for Gaussian process regression, and mixture reduction. I have applied my research to filtering and smoothing problems for non-linear and non-Gaussian state space models and (extended) target tracking using random matrices. Examples of such applications to real world problems include vehicle localization using GPS and inertial sensors, tracking bicycles using video cameras, and indoor localization using ultra-wideband sensors.

During the first year of my PhD research, I worked on optimal control problems and specifically optimal path tracking for robots.

A list of my publications can be found here.