Comprehensive survival analysis of lane-changing duration. (September 2021)

• Record Type: Journal Article
• Authors: Li, Yang; Li, Linbo; Ni, Daiheng; Zhang, Yue
• Is Part Of: Measurement. Volume 182(2021)
• Publication Date: 2021-09
• Subjects: Lane-changing duration -- Lane-changing behavior -- Survival analysis -- HighD dataset; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2021.109707
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: Lane-changing trajectories are extracted from the HighD dataset. Both comparative survival univariate and regression analysis has been conducted. The Generalized Gamma function has similar performance with non-parametric methods. LoglogisticAFT exhibits more credible results than other regression models. Abstract: Lane-changing duration (LCD) measures the total time it takes for a vehicle to travel from the current lane to the target lane. However, there is still a paucity research on LCD up to now. In this paper, we present a comprehensive analysis of LCD from the perspective of survival analysis. Naturalistic vehicle trajectory HighD dataset is employed in this paper, which contains of 16.5 h of measurement and over 11, 000 vehicles. Both comparative univariate and regression analysis has been conducted to research the characteristic of the whole survival function and the influencing factors of LCD. Results indicate that Generalized Gamma distribution has high degree of coincidence with the non-parametric method in estimating the survival function, and LoglogisticAFT model exhibits more credible results than other regression models. Furthermore, the results and modeling implications have been discussed. We hope this paper could contribute to our further understanding of LCD and LC behaviors.