The soil road is composed of powder and It is difficult to determine the driving state of unmanned vehicles because the road surface is deformed when the road surface is contacted with wheel. In order to secure driving performance, stability and efficiency of unmanned vehicles, it is necessary to establish a contact analysis database between wheels and road surfaces. Soil and sand properties are affected by environment, such as weather and humidity, so that property changes can easily occur. Because of soil and sand characteristics, to build a database by performing experiments on real land surfaces is not efficient. In simulation, since there is no effect on the surrounding environment, it is possible to build and efficient database. Therefore it is necessary to study wheel and soil road contact analysis through computer simulation.
The sandy road is modeled based on discrete element method and Full-Car is modeled based on multi-body dynamics. After calculating the contact force between the particle and particle, the particle and rigid body, the discrete element method applies the Velocity verlet integrator and the multibody dynamics is calculated using the 4th runge-kutta integrator.
In this study, wheel test bed experiments is performed to conduct contact studies between wheels and sandy road. Wheel test bed simulation is verified by comparison experiment of drawbar pull and max sinkage. Based on the wheel test bed experiments and simulation, Coast down driving simulation of full-car is performed and compared the driving distance, average deceleration of full-car and drawbar pull of wheels according to the Initial velocity. The simulation is expected to establish and efficient database of contact analysis between wheels and sandy road.