2d soil-bucket simulation using material point method for efficient earthwork modeling

Abstract

Since earthwork is quite repetitious, there has been a lot of research towards developing autonomous excavators. To automated earthwork equipment such as excavator and dozer, task planning (i.e. task sequence and location) play a critical role to execute automated earthwork as well as to increase work productivity. Soil-bucket interaction must be simulated to generate the task planning of an automated equipment, and, in this study, MPM (Material Point Method) were used to perform two-dimensional soil simulations. The strategy of determining the Lamé parameters that best show soil deformation was applied to the study when excavating with excavators. A game engine was used to test 15 different parameter combinations to visually check the changes in soil. It is expected that implementing more realistic excavation simulations can leads to increase the efficiency of automated excavators. More realistic excavation simulations should improve the effectiveness of autonomous excavators, according to the researchers.