Automated Methods for Converting a Non Real-time Cartesian Multi-body Vehicle Dynamics Model to a Real-time Recursive Model 2006-01-1165

The United States Army Research, Development, and Engineering Command’s Tank Automotive Research, Development and Engineering Center (U.S. Army RDECOM-TARDEC) laboratories is seeking to advance modeling and simulation methods used for predicting the performance of ground vehicles. TARDEC typically generates non-real-time models of its vehicles using DADS [1]: a general purpose commercial, multi-body software package based on a Cartesian coordinate formulation. TARDEC also currently uses SimCreator [2], [3] to develop real-time multi-body vehicle models. SimCreator uses recursive techniques to perform the simulations in real-time. The goal of the study presented here was to develop rapid conversion methods for translating models of DADS and other commercial multi-body software packages into SimCreator models. A procedure that can be automated was developed to convert a DADS model of a High-Mobility Multipurpose Wheeled Vehicle (HMMVW) to a SimCreator model. The vehicle model consisted of the chassis base body and the 4 wheels connected to it with double A-Arm suspensions and steering constraints. This model with 2 closed kinematic loops at each suspension was preprocessed to determine the best joints to be used as constraints (cut joints). The Cartesian joint coordinates of the DADS models were converted to relative joint coordinates used by SimCreator. Bodies, joints, and force elements were then converted using the components from SimCreator’s multi-body dynamics library. In place of the radial spring tire model used in the DADS simulation, a multi-disc tire model was developed to be used with SimCreator. The vehicle running over 8-inch, 6-inch, and 4-inch half-round obstacles at 10 mph was then simulated using the SimCreator model and the results were compared with field test results and similar simulations of the DADS model. Review of the results showed that the SimCreator results closely resembled the test data and was typically more accurate than the DADS results. Since, the multi-body effects of SimCreator and DADS should be generating the same results, it is believed that the differences between the results are due to the difference in the tire models.