Improving safety through online driver workload assessment

The continuous increasing number of car embedded systems requires to wonder about the compatibility of the use of such systems while the driving. The richness of their interface, their complexity and their numbers need to ensure that their use doesn´t exceed the driver capacity. To bring elements of answer to this question, this paper presents a methodology for the development of a real-time workload estimator. This methodology is based on the simultaneous analysis of objective and subjective data by the mean of an exploratory multivariate approach. Objective data are composed of driving performance indicators while subjective data are results from subjective workload assessments. The data were collected during an experimentation set on a driving situation which was designed to induce several levels of task demand.