Identification and Comparative Safety Analysis of Overtaking Patterns on Two-Lane Highways: A Simulator Study With Novice Drivers

TRANSPORTATION RESEARCH RECORD(2022)

引用 0|浏览1
暂无评分
摘要
Overtaking on two-lane highways is one of the riskiest driving maneuvers. However, safety analysis technologies for overtaking patterns and trajectories require development. In this simulator study, 328 overtaking trajectories were collected from 52 novice drivers to analyze difference in overtaking safety performances on two-lane highways under different traffic conditions. Two analyses were conducted: Analysis I focused on structural trajectory differences and classified the trajectory sample into two patterns-"ladder" (LAT) and "triangle" (TRT). In Analysis II, a comparative safety evaluation method was established based on data envelopment analysis to compare overtaking pattern safety; a factor analysis of the safety performance was undertaken by means of ordinary regression models. We found that TRT had a lower spatial consumption and smoother trajectory than LAT, but drivers overtaking using TRT were exposed to higher risk than LAT. Certain environmental and human factors had an influence on the safety levels of the patterns: both deteriorated when drivers were surrounded by vehicles driving at higher speeds, and improved with longer passing sight distances. However, safety levels when overtaking trucks were significantly worse than overtaking cars in the LAT pattern-the opposite was found for TRT. Female drivers' safety performance was superior in the LAT pattern; no significant gender difference were found with TRT. A longer following distance was found to ameliorate the safety of the TRT pattern, with no effect on LAT. These findings aid understanding of overtaking maneuvers on two-lane highways, and provide recommendations for advanced assistance driving and automatic driving systems.
更多
查看译文
关键词
two-lane highway, overtaking maneuver, safety, behavioral safety analysis and factor analysis
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要