Citation

Ma K, Wang H, Zuo Z, Hou Y, Li X, Jiang R. Transp. Res. C Emerg. Technol. 2022; 145: e103927.

Copyright

(Copyright © 2022, Elsevier Publishing)

DOI

10.1016/j.trc.2022.103927

PMID

unavailable

Abstract

Recent literature shows that automated vehicle (AV) significantly impacts traffic flow stability. However, as a widely used longitudinal AV technology, the Adaptive Cruise Control (ACC) system still has several unknown features that lack field data for modeling. We ran a series of novel tests with two ACC vehicles to simulate the behavior of the AV platoon in the field experiment. We designed a linear car-following controller in the upper layer system and tested the string stability of the ACC vehicle platoon. Data is collected through the vehicle's GPS sensor and upper controller. These data can help us estimate the feedback delay in the upper layer and parasitic lag in the lower layer. The data analysis results specify the range of different time delays and further optimize the dynamic response model in the lower layer. Then we propose an improved longitudinal model based on a first-order inertia system with a pure time delay in the dynamic response process. The experimental results show that the improved longitudinal model is twenty percent more accurate than the traditional first-order inertia model. Finally, we analyze the string stability of the improved longitudinal model. The results indicate that the improved longitudinal model has more strict stability conditions than the traditional model. Although the current feedback delay can satisfy the string stability criterion, the parasitic lag in the dynamics response process cannot satisfy the stability criterion. In this study, a customized ACC car-following controller is implemented in the field experiments, and thus all hypotheses and analysis results are validated by field data.

Language: en

Keywords

Adaptive cruise control; Automated vehicle; Field experiment; String stability