EAGER: Connected and Automated Vehicle Assessment Platform Using a Crowdsourced Cyber-Physical Reality

EAGER：使用众包网络物理现实的互联自动化车辆评估平台

• 批准号: 1844238
• 负责人: Joyoung Lee
• 金额: $ 30万
• 依托单位: New Jersey Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-09-15 至 2021-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1844238&HistoricalAwards=false
• 关键词: EAGER; Connected; Automated; Vehicle; Assessment

This EArly-concept Grant for Experimental Research (EAGER) project will produce an innovative assessment platform for the evaluation of the impacts of Connected and Automated Vehicles (CAVs) on individuals including drivers, passengers and pedestrians. Since its inception in the early 2000s, CAV technology has been expected to revolutionize the way a vehicle is controlled, potentially having significant implications on safety and efficiency of transportation systems. Existing evaluation approaches for CAVs heavily rely on computer simulations, which are insufficient for capturing the impact on human beings who will have various cognitive and behavioral reactions in response to CAVs. Missing this link, CAVs cannot be deployed on the road. By integrating robotics, 3-D printing, wireless network, traffic sensing and crowdsourcing technologies, the assessment platform that will be built through this EAGER project will reform the existing evaluation paradigm relying on simulations. Combining this platform with CAVs can create a positive feedback loop between design, development, and assessment of CAVs and thus enable its on-road deployment. This research primarily focuses on the assessment of CAV impacts on the individuals who are 1) passengers of CAVs, 2) drivers in the vicinity of CAVs, or 3) pedestrians around CAVs. To seamlessly capture their cognitive (e.g., safety awareness, degree of comfort) and behavioral reactions (e.g., steering maneuvers, accelerating or decelerating activities), this project develops a novel assessment platform utilizing a crowdsourced cyber-physical reality to realize the high-fidelity evaluations of CAVs. Using visual and force feedback, human involvement as drivers, passengers, and pedestrians could be assessed through cyber-physical reality. This EAGER project will address a number of technical challenges including: 1) how to assure that the platform's traffic dynamics would well represent the real-world situations; 2) how to seamlessly connect the human testers with miniature environment through virtual reality; and 3) how to design experimental scenarios to investigate a wide variety of potential problems in CAV technologies. The project will tackle the challenges by integrating robotics, 3D-printing, traffic engineering, and crowdsource technologies. The platform will be validated by comparing the quantitative measures collected from the proposed platform with those predicted by theories and sampled from real world traffic. It is also expected that demonstrations of the built platform will be widely conducted and disseminated at the end of the project.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概念实验研究资助（EAGER）项目将产生一个创新的评估平台，用于评估联网和自动驾驶车辆（CAV）对驾驶员、乘客和行人等个人的影响。自21世纪初问世以来，CAV技术一直被认为将彻底改变车辆的控制方式，可能对运输系统的安全性和效率产生重大影响。现有的评估方法在很大程度上依赖于计算机模拟，这是不足以捕捉的影响，人类将有各种认知和行为反应，以响应CAV。缺少这一环节，CAV就无法在公路上部署。通过整合机器人、3D打印、无线网络、交通传感和众包技术，将通过EAGER项目构建的评估平台将改革现有依赖模拟的评估范式。将该平台与CAV相结合，可以在CAV的设计、开发和评估之间建立一个积极的反馈循环，从而实现其道路部署。本研究主要集中于评估CAV对以下个体的影响：1）CAV的乘客，2）CAV附近的驾驶员，或3）CAV周围的行人。为了无缝地捕捉他们的认知（例如，安全意识、舒适度）和行为反应（例如，转向操纵，加速或减速活动），该项目开发了一个新的评估平台，利用众包的网络物理现实来实现CAV的高保真评估。使用视觉和力反馈，可以通过网络物理现实评估驾驶员，乘客和行人的人类参与。EAGER项目将解决一系列技术挑战，包括：1）如何确保平台的交通动态能够很好地代表真实世界的情况; 2）如何通过虚拟现实将人类测试人员与微型环境无缝连接;以及3）如何设计实验场景以调查CAV技术中的各种潜在问题。该项目将通过整合机器人、3D打印、交通工程和众包技术来应对挑战。将通过比较从所提出的平台收集的定量测量与理论预测的和从真实的世界交通中采样的定量测量来验证该平台。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Virtual Guide Dog: Next-generation pedestrian signal for the visually impaired

虚拟导盲犬：为视障人士提供下一代行人信号灯

• DOI: 10.1177/1687814019883096
• 发表时间: 2020
• 期刊: Advances in Mechanical Engineering
• 影响因子: 2.1
• 作者: Zhong, Zijia;Lee, Joyoung
• 通讯作者: Lee, Joyoung

Traffic Flow Characteristics and Lane Use Strategies for Connected and Automated Vehicles in Mixed Traffic Conditions

• DOI: 10.1155/2021/8816540
• 发表时间: 2021-01
• 期刊: Journal of Advanced Transportation
• 影响因子: 2.3
• 作者: Zijia Zhong;Joyoung Lee;Liuhui Zhao

Development and Evaluation of Cooperative Intersection Management Algorithm under Connected and Automated Vehicles Environment

网联自动化车辆环境下协同交叉口管理算法的开发与评估

• DOI: 10.1177/0361198121994580
• 发表时间: 2021
• 期刊: Transportation Research Record: Journal of the Transportation Research Board
• 作者: Gutesa, Slobodan;Lee, Joyoung;Besenski, Dejan
• 通讯作者: Besenski, Dejan

A Deep Reinforcement Learning Network for Traffic Light Cycle Control

• DOI: 10.1109/tvt.2018.2890726
• 发表时间: 2019-02-01
• 期刊: IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
• 影响因子: 6.8
• 作者: Liang, Xiaoyuan;Du, Xunsheng;Han, Zhu
• 通讯作者: Han, Zhu

High-Fidelity Teleoperated Scaled Vehicles for Research and Development of Intelligent Transportation Technologies

用于智能交通技术研发的高保真遥控比例车

• DOI: 10.1115/dscc2020-3173
• 发表时间: 2020
• 期刊: ASME 2020 Dynamic Systems and Control Conference
• 作者: Wang, Cong;Wang, Bohan;Zhao, Leidi;Maranon, Carlos;Goswamy, Nishaant;Lee, Jo Young;Wang, Guiling
• 通讯作者: Wang, Guiling