Mcity 2: An Integrated Automated Testbed for Autonomous Transportation Research

NSF民用基础设施系统项目将为自动交通研究创建世界上第一个全自动增强现实测试设施。互联和自动驾驶汽车（CAV）技术具有显著提高安全性、机动性和可持续性的潜力，但也在技术和社会层面（包括控制、传感、通信、人在回路等）提出了巨大挑战。大多数研究想法都是通过有限的开放数据集进行模拟测试的。物理测试设施和大规模真实世界数据的获取受到限制，严重阻碍了科学研究和技术开发的进展。 该项目将利用密歇根大学现有的Mcity测试设施-世界上第一个专门建造的CAV试验场-通过将物理测试轨道与软件模拟环境集成在一起，以创建增强现实CAV试验台，该试验台将通过基于云的访问提供给全国的学术研究人员。该项目将为全国范围内的合作发现和集体努力提供手段，以加速CAV技术的研究和劳动力开发。 该项目还将通过为研究界，特别是那些来自服务不足社区的资源较少的研究界提供更容易获得顶级研究基础设施的机会，大大扩大参与。 Mcity增强现实测试平台集成了三个组件：（1）包括基础设施和CAV车队的物理测试设施，（2）收集和共享21个交叉路口近实时交通信息的移动数据中心，（3）增强自然潜水模拟器。 该试验台将是自动化的，因此研究人员可以配置和控制测试设施基础设施（例如，交通信号灯、人行横道按钮、铁路交叉臂等）并使用基于Web的图形用户界面构建测试场景。移动数据中心将向研究界公开提供安阿伯生活实验室中路边单元（RSU）广播的传感器数据共享消息（SDSM）。这些数据将用于开发自然驾驶环境（NDE）模拟器，用户可以在物理Mcity测试设施部署之前测试他们的算法，混合了真实的和虚拟的道路使用者。所有测试数据都将存档在云数据库中，以实现安全访问和数据/方法共享，用于研究和教育目的。将成立一个由政府、学术界和工业界成员组成的技术咨询委员会，以确保公平和有效地利用这一研究基础设施。 每年都计划开展教育和推广活动，以加强学术界与工业界的合作伙伴关系，并为新兴的CAV行业发展劳动力。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。

This NSF Civil Infrastructure Systems project will create the world’s first fully automated augmented-reality test facility for autonomous transportation research. Connected and automated vehicle (CAV) technologies have the potential to significantly improve safety, mobility, and sustainability, but also pose great challenges in technical and social dimensions including control, sensing, communication, human-in-the-loop, etc. Although significant research effort has been dedicated to the research and development of CAV technologies, most research ideas have been tested using simulation only with limited open datasets. Access to physical test facilities and large-scale real-world data is limited, severely hindering the progress of scientific research and technology development. This project will leverage the University of Michigan’s existing Mcity Test Facility—the world’s first purpose-built CAV proving ground—by integrating the physical test track with a software simulation environment to create an augmented-reality CAV testbed that would be made available to academic researchers nationwide through cloud-based access. This project will provide the means for nation-wide collaborative discovery and collective effort to accelerate the research and workforce development of CAV technologies. The project will also significantly broaden participation by providing easier access to top-tier research infrastructure for the research community, especially those with less resources from under-served communities. The Mcity augmented reality testbed integrates three components: (1)a physical test facility including infrastructure and CAV fleet, (2) a mobility data center that collects and shares near-real-time traffic information from 21 intersections, (3) an augmented naturalistic diving simulator. This testbed will be automated so researchers can configure and control the test facility infrastructure (e.g., traffic lights, crosswalk buttons, rail-crossing arms, etc.) and build test scenarios using a web-based graphical user interface. The Mobility Data Center will make available sensor data sharing messages (SDSMs) broadcast by roadside units (RSUs) in the Ann Arbor Living Laboratory publicly to the research community. Such data will be used to develop a naturalistic driving environment (NDE) simulator that users can test their algorithms before deployment at the physical Mcity Test Facility, with a blend of real and virtual fellow road users. All test data will be archived in a cloud database to enable secure access and data/method sharing for research and education purposes. A Technical Advisory Committee, composed of members of government, academia, and industry, will be established to ensure equitable and effective usage of this research infrastructure. Education and outreach activities are planned annually to strengthen academic-industry partnerships and workforce development for the emerging CAV industry.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.