A study on Autonomous Vehicles’ Driving Behavior Based on Open Road Test Data

In this paper, the Autonomous Vehicles (AVs)’ actual driving performance in open road test and its impacts on traffic were studied. We used free flow speed, minimum headway, maximum acceleration and deceleration as characteristic parameters for AVs’ driving behavior. And based on the matching results of trajectory data and road network, the above parameters were extracted. The analysis of the above characteristic parameters showed that the values of the free flow speed, minimum headway, and maximum acceleration and deceleration of AVs in the test cannot be expected better than the corresponding values of Human-Driven Vehicles (HVs) like many previous relevant studies did. Actually, according to the extracted values and dispersions of the behavior parameters, the AVs even perform worse than HVs. As a result, the expected significant improvements in the operational efficiency and safety of the traffic causing by AVs did not occur according to the simulation. There were even signs of deterioration. The results of this paper will help people re-examine the behavior parameter settings of AVs in study and provide reference for enterprises and governments to improve the behavior of AVs from the perspective of traffic system efficiency and safety.