Numerical analysis and experimental verification of vehicle trajectories

Abstract
The paper presents research results of a study, in which computational mechanics was utilized to predict vehicle trajectories upon traversing standard Florida DOT street curbs. Computational analysis was performed using LS-DYNA non-linear, finite element computer code with two public domain, finite element models of motor vehicles: Ford Festiva and Ford Taurus. Shock absorbers were modeled using discrete spring and damper elements. Connections for the modifie suspension systems were carefully designed to assure proper range of motion for the suspension models. Inertia properties of the actual vehicles were collected using tilt-table tests and were used for LS-DYNA vehicle models.

Full-scale trajectory tests have been performed at Texas Transportation Institute to validate the numerical models and predictions from computational mechanics. Experiments were conducted for Ford Festiva and Ford Taurus, both for two values of approach angle: 15 and 90 degrees, with impact velocity of 45 mph. Experimental data including accelerations, displacements and overall vehicles behavior were collected by high-speed video cameras and have e been compared with numerical results. Verification results indicated a good correlation between computational analysis and full-scale test data. The study also underlined a strong dependence of properly modeled suspension and tires on resulting vehicle trajectories.