Paper presented at Driving Simulator Conference 2016, 7–9 September 2016, Paris, France. Authored by van Doornik, J., Brems, W., de Vries, E.J.H., and Wiedemann, J.
Latency in different components has always been a big challenge in the design of driving simulators, when trying to simulate the most realistic driving experience in a real car. As simulators are subject to the laws of physics, there will always be some latency in the system. However, many simulator users have validated vehicle models, accurately representing real car behaviour ; and they want to be able to use these on the simulator without having to spend time, effort, money and other resources on a re-validation for simulator work. This paper presents a systematic approach to synchronize and reduce simulator latency using prediction algorithms, without affecting the vehicle model and its behaviour. The final goal of this research is to create a simulator with “zero perceived latency” for the driver. At the same time the maximum induced prediction errors are specified. The resulting algorithms are implemented in two different driving simulators. The effectiveness and the quality of the prediction algorithms are tested in different driving clinics. Results show that the implemented algorithms are robust and reliable. Furthermore the results indicate that also small differences in latency are subjectively perceived by the drivers, and can be objectively measured to some extent.