Highway traffic data: macroscopic, microscopic and criticality analysis for capturing relevant traffic scenarios and traffic modeling based on the highD data set

Abstract: This work provides a comprehensive analysis on naturalistic driving behavior for highways based on the highD data set. Two thematic fields are considered. First, some macroscopic and microscopic traffic statistics are provided. These include the traffic flow rate and the traffic density, as well as velocity, acceleration and distance distributions. Additionally, the dependencies to each other are examined and compared to related work. The second part investigates the distributions of criticality measures. The Time-To-Collision, Time-Headway and a third measure, which couples both, are analyzed. These measures are also combined with other indicators. Scenarios, in which these measures reach a critical level, are separately discussed. The results are compared to related work as well. The two main contributions of this work can be stated as follows. First, the analysis on the criticality measures can be used to find suitable thresholds for rare traffic scenarios. Second, the statistics provided in this work can also be utilized for traffic modeling, for example in simulation environments.