TR2019-145

Bayesian Learning of Tire Friction with Automotive-Grade Sensors by Gaussian-Process State-Space Models


    •  Berntorp, K., Hiroaki, K., "Bayesian Learning of Tire Friction with Automotive-Grade Sensors by Gaussian-Process State-Space Models", IEEE Conference on Decision and Control (CDC), December 2019.
      BibTeX Download PDF
      • @inproceedings{Berntorp2019dec,
      • author = {Berntorp, Karl and Hiroaki, Kitano},
      • title = {Bayesian Learning of Tire Friction with Automotive-Grade Sensors by Gaussian-Process State-Space Models},
      • booktitle = {IEEE Conference on Decision and Control (CDC)},
      • year = 2019,
      • month = dec,
      • url = {https://www.merl.com/publications/TR2019-145}
      • }
  • MERL Contact:
  • Research Areas:

    Control, Machine Learning, Signal Processing


The friction dependence between tire and road is highly nonlinear and varies heavily between different surfaces. Knowledge of the tire friction is important for real-time vehicle control, but difficult to estimate with automotive-grade sensors. Based on recent advances in particle filtering and Markov chain Monte-Carlo methods, we propose a batch method for identifying the tire friction as a function of the wheel slip. The unknown function mapping the wheel slip to tire friction is modeled as a Gaussian process (GP) that is included in a dynamic vehicle model relating the GP to the vehicle state. The method is able to efficiently learn the tire friction using only wheel-speed, steering-wheel angle, and inertial automotivegrade sensors. We illustrate the efficacy of the method using several experimental data sets obtained on a snow-covered road.