RUI: Behavior-Based Stochastic Traffic Flow Modeling for Intersection Safety Improvement

RUI：基于行为的随机交通流建模，用于改善交叉口安全

• 批准号: 1536277
• 负责人: Xinkai Wu
• 金额: $ 20万
• 依托单位: Cal Poly Pomona Foundation, Inc.
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-10-01 至 2019-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1536277&HistoricalAwards=false
• 关键词: RUI; Behavior; Based; Stochastic; Traffic

This research will investigate driver behavior at signalized intersections and derive sophisticated traffic flow models aimed at intersection safety improvement. Intersection safety has long been a national concern, partly due to the lack of understanding of complicated driving behaviors, especially decision-making mechanisms present when drivers face signal phase changes. This research tackles this issue by conducting a comprehensive investigation of driver decision-making at signalized intersections and, using insights from these investigations, developing a stochastic traffic flow model for such interrupted flow situations. Through the integration of the developed traffic flow model with connected vehicle technologies (specifically, V2X), the developed traffic flow model can be used to quantify the safety performance of signalized intersections, identify emerging hazardous situations, and contribute to the development of driver-assistance and intersection-accident-avoidance technologies. Furthermore, the outcomes from this research will create opportunities for students, particularly those from underrepresented groups, through educational and research opportunities in transportation engineering. It will also provide opportunities for college students, traffic engineers, and K-12 faculty and students through curriculum development, educational modules, and creation of pathways between high school and college, as well as college and graduate school.The objectives of this research include: 1) investigating complicated driving behaviors and the inner mechanisms of drivers' decision-making at signalized intersections through the analysis of vehicular trajectory data extracted from video images and 2) developing a stochastic traffic flow model to describe complicated driving behaviors and predict potential traffic conflicts at signalized intersections by considering the stochastic nature of drivers decision-making when facing signal phase changes under varying circumstances. This research will contribute to the theoretical development of traffic flow models. These traffic flow models will be able to describe complicated driving behaviors and estimate traffic conflicts at signalized intersections, both of which are absent from most other traffic flow models. Furthermore, this research is expected to contribute significantly to the improvement of intersection safety. This research will build a foundation for the future development of dynamic systems for alerting drivers of emerging hazards and helping to avoid intersection accidents.

本研究将调查信号交叉口的驾驶员行为，并推导出复杂的交通流模型，以提高交叉口的安全性。长期以来，交叉口安全一直是全国关注的问题，部分原因是缺乏对复杂驾驶行为的理解，特别是当驾驶员面对信号相位变化时所呈现的决策机制。本研究通过对信号交叉口的驾驶员决策进行全面调查来解决这一问题，并利用这些调查得出的见解，为这种中断的交通情况开发了一个随机交通流模型。通过将已开发的交通流模型与车联网技术（特别是V2X）相结合，所开发的交通流模型可用于量化信号交叉口的安全性能，识别新出现的危险情况，并有助于驾驶员辅助和交叉口事故避免技术的发展。此外，这项研究的结果将为学生创造机会，特别是那些来自代表性不足的群体，通过交通工程的教育和研究机会。它还将通过课程开发、教育模块、高中和大学、大学和研究生院之间的通道的创建，为大学生、交通工程师、K-12教师和学生提供机会。本研究的目的包括：1)通过分析视频图像中提取的车辆轨迹数据，研究信号交叉口复杂驾驶行为及其决策的内在机制；2)考虑不同情况下信号相位变化下驾驶员决策的随机性，建立随机交通流模型，描述信号交叉口复杂驾驶行为并预测潜在的交通冲突。本研究将有助于交通流模型的理论发展。这些交通流模型将能够描述复杂的驾驶行为和估计信号交叉口的交通冲突，这是大多数其他交通流模型所缺乏的。此外，本研究可望对提高交叉口的安全性有重要的贡献。这项研究将为未来动态系统的发展奠定基础，以提醒驾驶员注意新出现的危险，并帮助避免交叉事故。