Highlights New vehicle–snow interaction data for combined-slip conditions were obtained. Stochastic models were developed for calibration and validation. Four metrics were used to assess the quality of validation with good results.
Abstract General operations of a vehicle involve simultaneous slip in the longitudinal and lateral directions, the combination of which is much more complicated than purely longitudinal or lateral motions. During vehicle–snow interactions, additional complexities arise due to uncertainties of snow material properties and of interfacial properties between the tire and snow, calling for the stochastic modeling of the interactions to validate the model. For validation, a statistical framework was formed with several components: a deterministic, physically-based tire–snow interaction model, a stochastic metamodel based on the physical model, a statistical model for calibration, prediction using the models, validation metrics, and new test data using an instrumented vehicle. The longitudinal and lateral drawbar pulls, and the torque and overturning moment, were used simultaneously to calibrate model parameters for a front and a rear tire. Four local and global validation metrics and extensive summary statistics were used to assess the quality of the models, with good results.
Vehicle–snow interaction: Testing, modeling and validation for combined longitudinal and lateral slip
Journal of Terramechanics ; 55 ; 1-16
2014-05-17
16 pages
Article (Journal)
Electronic Resource
English
Snow , Drawbar pull , Traction , Slip angle , Validation metrics , Calibration , Bayesian , Metamodel , Stochastic , Gaussian process
Automotive engineering | 2014
|Vehicle-wet snow interaction: Testing, modeling and validation
Elsevier | 2016
|Vehicle-wet snow interaction: Testing, modeling and validation
Online Contents | 2016
|