Soil Deformation Model for Efficient Simulation of Off-Road Vehicles

An efficient and accurate soil deformation model is crucial for dynamic simulations of off-road vehicles. This paper presents a soil deformation model that specifically focuses on the soil cutting (bulldozing) process. Two different methods are developed to achieve this goal. The first method is based on deep learning and utilizes the discrete element method (DEM) to acquire training data. The second method relies on soil mechanics semi-empirical theories. To validate the performance of both methods in predicting soil cutting force and soil deformation, DEM simulation results are employed. The predictions generated by both methods demonstrate strong agreement with the DEM simulation results.