Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


State and friction estimation for 1:43 RC cars


D. Stettler

Semester Thesis, FS15 (10410)

In order to improve the state estimation of the autonomos RC racing project ORCA at the Automatic Control Laboratory at ETH Zurich, three algorithms are presented, which are always built upon the previous ones. Firstly, an extended Kalman filter is derived for the given car model. Using the model of the car and the known control inputs, the state is estimated more accurately. Compared to the current estimator, an additional advantage of the implemented Kalman filter is, that it can be extended to solve additional problems, which is done in the next two steps. In a second step, a delay compensation algorithm is developed. Since the vision system requires some time to get the position and orientation out of an image, the data received by the controller does not match the actual state anymore. Here, a delay of about 30 ms was experimentally determined. Using an extended state, it is achieved that the Kalman filter can estimate the current state, based on the delayed measurement from the camera system. Lastly, to further improve the estimation and to be adaptive to changes in the track underground or the tire condition, the friction parameters between the car's wheel and the track are estimated online using an extended state. All three algorithms are first tested in MATLAB using recorded data and then implemented to the real system, where the results are evaluated and a stable working platform is generated.

Supervisors: Alexander Liniger, John Lygeros


Type of Publication:

(13)Semester/Bachelor Thesis

J. Lygeros

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2015:IFA_5279
Permanent link