Virtual model correlation and its validation using a dynamic driver simulator 2020-36-0179

In automotive industry launches, vehicle dynamics play a key role in the success of each project, as it must make that product recognizable as belonging to a particular automaker and must also reflect what the customer expects from the product itself. So, during the initial phases of the project, a study of reference vehicles begins experimentally, building a robust basis for the definition of performance targets, using not only objective data but also a subjective evaluation, resulting in a technical specification document with most relevant points. Thus, with a known and measured reference vehicle, a virtual model should be created to serve as basis for the development of the new vehicle. In this first moment, the reference model must faithfully represent the values measured on the track, ensuring a robust comparison basis for the future development. The numerical-experimental correlation occurs at several levels, from the correlation of suspension and steering kinematics and compliance to full vehicle’s dynamic behavior, in this case, lateral dynamics. With a correlated multibody model, two steps are taken, converting it to the real-time environment and, in other to derive the model that will represent the new vehicle. With this new model, the same maneuvers will be performed to compare and from that point on, new solutions can be explored to reach the project targets, after this phase the model of the new vehicle is also converted to the real-time environment, thus allowing the simulation to occur in real-time on the dynamic simulator. Once in the simulator, the driver will evaluate the reference model and adjust the motion cueing, that is a platform response algorithm to adjust the driver's subjective perception of the model since objectively it responds according to the physical measurement, the same motion cueing must be applied on the new vehicle model.