TR2017-093
Particle Gibbs with Ancestor Sampling for Identification of Tire-Friction Parameters
-
- , "Particle Gibbs with Ancestor Sampling for Identification of Tire-Friction Parameters", World Congress of the International Federation of Automatic Control (IFAC), DOI: 10.1016/j.ifacol.2017.08.2585, July 2017, vol. 50, pp. 14849-14854.BibTeX TR2017-093 PDF
- @inproceedings{Berntorp2017jul,
- author = {Berntorp, Karl and Di Cairano, Stefano},
- title = {Particle Gibbs with Ancestor Sampling for Identification of Tire-Friction Parameters},
- booktitle = {World Congress of the International Federation of Automatic Control (IFAC)},
- year = 2017,
- volume = 50,
- number = 1,
- pages = {14849--14854},
- month = jul,
- publisher = {Elsevier},
- doi = {10.1016/j.ifacol.2017.08.2585},
- url = {https://www.merl.com/publications/TR2017-093}
- }
- , "Particle Gibbs with Ancestor Sampling for Identification of Tire-Friction Parameters", World Congress of the International Federation of Automatic Control (IFAC), DOI: 10.1016/j.ifacol.2017.08.2585, July 2017, vol. 50, pp. 14849-14854.
-
MERL Contact:
-
Research Areas:
Abstract:
Particle Gibbs with Ancestor Sampling (PGAS) is a particle Markov chain Monte Carlo method (PMCMC) for Bayesian inference and learning. PGAS conditions on a referencestate trajectory in the underlying particle filter using ancestor sampling. In this paper, we leverage PGAS for identification of cornering-stiffness parameters in road vehicles only using production-grade sensors. The cornering-stiffness parameters are essential for describing the motion of the vehicle. We show how PGAS can be adapted to efficiently learn the stiffness parameters by conditioning on the noise-input trajectory instead of the state trajectory. We verify on a three-minute long experimental test drive that our method correctly identifies the tire-stiffness parameters.