Traction is an important performance requirement for a tire and is the force that propels a vehicle forward. Accurate traction prediction has value in reducing tire development cycle time, improving mechanistic understanding of tire performance and improving tire performance by achieving good soil traction while minimizing soil compaction. On a rigid surface like paved road, traction can be predicted using an appropriate tire-road friction model in a simulation of tire performance. However, traction on a flexible media like soil, snow, water, etc. is a more complex phenomenon. Available traction on soil is the sum of friction at tire-soil interface and the amount of soil shear strength extracted by the tire to propel itself.

We present the approach to predicting traction in soil by considering soil as a continuum in an Eulerian domain and the tire within a Lagrangian domain. Modeling soil has unique challenges due to its complex mechanical response and excessive deformation of the material. New internally developed constitutive equations for soil based on theory of plasticity with non-associated flow rule were used to improve the prediction of mechanical behavior of soil. The predicted and the measured mechanical response of soil are compared in triaxial loading conditions.

The resulting system of partial differential equations for a tire rolling on soil is solved using LS-DYNA, a commercial Finite Element solver and the new constitutive equations for soil implemented as a user material-subroutine. The contact forces at the soil-tire interface are resolved to obtain the traction forces on the tire. The measured and the prediction traction are compared for a rigid wheel rolling on soil, a plain tread agricultural tire rolling on soil and lugged agricultural tires rolling on soil. The results indicate that the new constitutive equations for soil and the continuum approach for modeling soil can successfully predict traction of a tire rolling on soil.