Collaborative Research: Modeling Human Driving Behavior and Response with Applications to Intelligent Agent-Based Traffic Flow Simulation

合作研究：利用基于智能代理的交通流模拟对人类驾驶行为和响应进行建模

• 批准号: 0527508
• 负责人: Lu Sun
• 金额: --
• 依托单位: Catholic University of America
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-01-01 至 2009-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0527508&HistoricalAwards=false
• 关键词: Collaborative; Research; Modeling; Human; Driving

Human decision-making in everyday vehicular traffic situations determines in large measure the properties of the traffic stream and the resulting quality of traffic service, which affect the reliability, safety and efficiency of transportation systems and the quality of urban life. This project puts forward a comprehensive, multidisciplinary research approach to characterize and model human cognitive driving behavior and subsequent response in traffic flow systems. Specifically, the dynamics of driver behavior, taken at the individual level and as part of a group, evolving over time and space will be systematically studied as a complex system. By developing behavior-based models of human decision-making in traffic situations and integrating the behavior models in computer simulation systems, the study addresses fundamental questions in traffic science and promises to improve prevailing understanding of traffic flow phenomena as well as the fidelity and reliability of the current state of the art of traffic flow simulation. The outcome of the project will reveal the interrelationship between microscopic elements of driver behavior (e.g. reaction times and other time lags, over-reaction, risk averse and risk seeking behaviors in car following and lane changing) and macroscopic traffic flow phenomena through analytical investigation and simulation, validated by field observation. The subject of the investigation entails considerable economic and social significance, with wide application in human cognitive science, as well as in traffic operations and management. The project also provides an excellent and unique opportunity for student exposure to cross-disciplinary research and helps build a diverse scientific workforce.

日常车辆交通状况的人类决策在很大程度上决定了交通流量的特性以及产生的交通质量，这会影响运输系统的可靠性，安全性和效率以及城市生活的质量。 该项目提出了一种全面的，多学科的研究方法，以表征和建模人类的认知驾驶行为和随后在交通流系统中的响应。具体而言，在单个级别和小组的一部分中采用的驾驶员行为的动力学，随着时间和空间的发展将被系统地研究为复杂的系统。通过在交通状况中开发基于行为的人类决策模型，并将行为模型整合到计算机模拟系统中，该研究解决了交通科学中的基本问题，并有望提高对交通流现象的普遍理解，以及交通流量模拟艺术现行状态的忠诚度和可靠性。该项目的结果将揭示驾驶员行为的显微镜元素（例如，反应时间和其他时间滞后，反应过度，反应，风险反对和风险寻求风险的行为）与泳道更改的行为和宏观交通流动现象通过分析和模拟，通过现场观察验证。调查的主题具有相当大的经济和社会意义，在人类认知科学以及交通运营和管理中的广泛应用。该项目还为学生接触跨学科研究提供了绝佳的独特机会，并有助于建立多样化的科学劳动力。