Transportation Expertise

Driver Modeling and Simulation

UMTRI researchers use computational modeling techniques to describe, explain, predict, and demonstrate driver behavior and performance. Modeling efforts are usually based on data collected in traditional human factors experiments in the simulator and on the road. Our models aim to extend the value of empirical results by making engineering predictions and attempting to explain phenomena related to basic driving and to driving in the presence of ITS systems. Discrete event simulation packages such as ProModel and Microsaint are used to generate stochastic predictions of driver performance. Research on the integration of cognitive modeling (QN models) with physical modeling (HUMOSIM) and task analysis (IMPRINT) as well as vehicle dynamics (Matlab) are part of this effort.