Headway Distribution Considering Vehicle Type Combinations

Abstract

To achieve more-accurate traffic simulation and autonomous vehicle control, increasing attention recently has been devoted to the analysis of car-following behavior considering leader–follower vehicle types. Firstly, on the basis of a vehicle trajectory data set, distance headway (DHW) and time headway (THW) distributions of leader–follower vehicle type combinations, including car following car (CC), car following truck (CT), truck following car (TC), and truck following truck (TT), were compared to verify their significant differences. Then best-fit distribution models of four combinations were identified in each scenario. Next, characteristics of DHW and THW of four combinations in different speed ranges were determined according to their best-fit distribution models. Finally, the transferability of the results was verified. Results show that DHW and THW of CC and CT can be modeled as log-logistic distributions, and DHW and THW best-fit models of TC and TT are gamma distributions. CF behavior depends more on the following vehicle type when the speed is less than 26 m/s, and the impact of the leading vehicle gradually occurs when the speed is higher than 26 m/s.