Talk of Thomas Schoen on "Nonlinear system identification and sensor fusion"

Speaker: Thomas Schoen (University of Linkoeping)
Time and Location: Thursday May 19, 11:00, Aula Magna

Title: Nonlinear system identification and sensor fusion

This talk consists of two parts, one more theoretical part on nonlinear
system identification and one more applied part on sensor fusion applications.

The system identification part is concerned with the parameter estimation of a general class of nonlinear dynamic systems in state-space form. More specifically, a Maximum Likelihood (ML) framework is employed and an Expectation Maximisation (EM) algorithm is derived to compute these ML estimates.
The Expectation (E) step involves solving a nonlinear state estimation problem, where the smoothed estimates of the states are required. This problem lends itself perfectly to the particle smoother, which provides arbitrarily good estimates. We will show how to identify a nontrivial Wiener model as an example of how the method can be used.

The second part of the talk is concerned with sensor fusion, which deals with the problem of how to make use of measurements from several different, often complementary, sensors in order to obtain better state estimates. This part of the talk is based on recent sensor fusion applications that we have been working on, including; indoor pose estimation of a human body (using inertial sensors and UWB), pose estimation of a helicopter using a map (using inertial sensors and a cameras), and real-time pose estimation of a helicopter in order to facilitate autonomous landing (using inertial sensors and cameras).


Contact: Giorgio Grisetti (