EAGER: Characterizing Driver Interactions with Emergency Vehicles

EAGER：描述驾驶员与紧急车辆的交互

• 批准号: 2332183
• 负责人: Robert Kluger
• 金额: $ 20.95万
• 依托单位: University of Louisville Research Foundation Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-10-01 至 2025-09-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2332183&HistoricalAwards=false
• 关键词: EAGER; Characterizing; Driver; Interactions; Emergency

This EArly-concept Grant for Exploratory Research (EAGER) funded project explores a new concept on how individual drivers behave during interactions with emergency vehicles (EV) in an experimental, virtual reality setting. Emergency vehicle response is critical to public health, and positive encounters with active emergency vehicles by drivers can both improve response time and reduce the risk of motor vehicle crashes. However, driver tendencies during these interactions are poorly understood, due to limited data availability and high dimensionality. This project designs and tests a novel approach that leverages data and insights from external sources (e.g., the SHRP2 Naturalistic Driving Study and MCity 2.0) to validate experimental models. The success of this exploratory effort is likely to unlock opportunities for deriving a library of EV encounters from virtual reality simulator systems and other datasets. Additionally, realistic models of behaviors have the potential to advance microsimulation of emergency encounters and enable the systematic study of management strategies. The common method could also be applicable to other rare driving scenarios. An adaptive scheme is developed and implemented in a virtual reality driving simulator experiment. This simulator emulates data during encounters with emergency vehicles in a controlled setting. The priors are updated with the result from microsimulation to identify scenarios offering the most information given the existing models. The collected data is then used to estimate drivers’ interactions with emergency vehicles including lane-changing behaviors, acceleration and braking, reaction time, awareness, and more. These elements are classified into reaction types by clustering of multivariate time series using the hidden Markov model. The validation involves a comparison of the developed models based on emulated data to features extracted from naturalistic driving data.This project is jointly funded by Civil Infrastructure Systems (CIS) and the Established Program to Stimulate Competitive Research (EPSCoR).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

EARLY概念探索性研究基金（EAGER）资助的项目探索了一个新的概念，即在实验性的虚拟现实环境中，驾驶员在与紧急车辆（EV）互动时的行为。紧急车辆响应对公共卫生至关重要，驾驶员积极主动地遇到紧急车辆既可以缩短响应时间，又可以降低机动车碰撞的风险。然而，在这些相互作用的驱动程序的趋势知之甚少，由于有限的数据和高维度。该项目设计并测试了一种新的方法，该方法利用了来自外部来源的数据和见解（例如，SHRP2自然驾驶研究和MCity 2.0）来验证实验模型。这项探索性工作的成功可能会为从虚拟现实模拟器系统和其他数据集中获得EV遭遇库提供机会。此外，现实的行为模型有可能推进微观模拟的紧急情况下，使管理策略的系统研究。该通用方法也可适用于其他罕见的驾驶场景。在虚拟现实驾驶模拟器实验中，开发并实现了自适应方案。该模拟器在受控设置中模拟遇到紧急车辆时的数据。先验的微观模拟的结果进行更新，以确定现有的模型提供最多的信息的情况下。然后，收集的数据用于估计驾驶员与紧急车辆的交互，包括变道行为、加速和制动、反应时间、意识等。这些元素被归类为反应类型的多变量时间序列的聚类使用隐马尔可夫模型。验证涉及基于模拟数据开发的模型与从自然驾驶数据中提取的特征的比较。该项目由民用基础设施系统（CIS）和刺激竞争研究的既定计划（EPSCoR）共同资助。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。