An Approach for the Estimation of Error Rates in Observing Critical Situations

Abstract

In this paper the observation quality of critical situations in road traffic is analyzed. Determining the frequency of critical situations is a well-established approach in road traffic safety studies. In this paper, a corresponding calculation scheme is deduced and applied to the problem of error rate calculation. In this context, the joint probability distribution is estimated by the joint frequency distribution of velocity difference and headway from a publicly available data set of the Intelligent Cruise Control Field Operational Test study. The mathematically predicted error rates are evaluated versus detection data of a state of the art video surveillance system applied to a real world test site.