Paper presented at Driving Simulator Conference 2015, 16–18 September 2015, Tübingen, Germany. Authored by Brems, W., van Doornik, J., de Vries, E.J.H., Wiedemann, J.

Driving simulators come in various sizes and are used for car development, driver training and much more; but can a mid-size simulator be used for subjective evaluation of vehicle dynamics in chassis development of normal cars? To answer this question a frequency response and latency analysis is conducted to find the performance characteristics of a Cruden simulator built up at AUDI AG. Describing functions, crosstalk and signal-to-noise ratios for the motion platform are evaluated in 6DoF, based on methods described in the AGARD-144 Advisory Report [Lea79]. For all 6DoF the system is limited in bandwidth by the first eigenfrequency, outside the frequency range of interest. Two different methods are used to evaluate frequency response for the force-feedback steering wheel, showing results with- and without human physics in the dynamic control loop. Visual latency is higher than the dynamic threshold for the motion platform and control loader, but still less than 30 ms using the fastest settings. There is not much room for improvement of the rendering software, as two-thirds of the visual latency comes from the projectors. These results show that the platform is suitable for subjective evaluation of vehicle dynamics ranging from 0-10 Hz.

Click here to access the source of the article.

‍