SaTC: TTP: Medium: Collaborative: Exposing and Mitigating Security/Safety Concerns of CAVs: A Holistic and Realistic Security Testing Platform for Emerging CAVs

联网和自动驾驶汽车（CAV）技术实现了实时信息共享和驾驶自动化，具有显著提高交通系统安全性和效率的潜力。然而，网络安全威胁可能会影响基础设施的运行效率和乘客的安全，对自动驾驶汽车的部署构成重大挑战。该合作项目开发了一种新的CAV测试平台，以有效和现实的方式解决评估CAV系统安全和安全问题的关键需求。CAV行业中的硬件制造商、软件开发人员和安全服务提供商都可以利用这样一个平台方便地全面测试他们的产品。此外，该测试平台可用于学术界和工业界的培训和教育，并促进工业界安全最佳实践和标准的发展。测试平台提供进攻性和防御性测试服务，涵盖自动驾驶汽车生态系统的三个部分：(1)交通基础设施；(2)互联车辆通信通道；(3)车载软件平台。它是第一个支持跨组件攻击和防御测试服务的CAV安全测试平台，旨在捕获三个关键组件之间的相互关系。该项目通过有效地结合优化、统计建模、机器学习、网络仿真、程序分析和模型检查等技术来开发测试支持，同时支持已知防御解决方案的创新使用，包括可信计算硬件、访问控制策略实施和异常检测。该项目在实际的CAV测试平台中实现和部署测试平台，并与行业合作伙伴合作，以便尽早采用。该奖项反映了美国国家科学基金会的法定使命，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。

Connected and Automated Vehicle (CAV) technologies enable real-time information sharing and driving automation, with the potential of significantly improving safety and efficiency of the transportation system. However, cyber-security threats may compromise the efficiency of infrastructure operations and the safety of passengers, posing a significant challenge for CAV deployment. This collaborative project develops a novel CAV testing platform to address the critical needs for assessing the security and safety concerns of the CAV system in an effective and realistic manner. Hardware manufacturers, software developers, and security service providers in the CAV industry can all leverage such a platform to conveniently and holistically test their products. Moreover, the testing platform can be used for training and education in both academia and industry, and facilitate the development of security best practices and standards in industry.The testing platform provides both offensive and defensive testing services covering three parts of the CAV ecosystem: (1) transportation infrastructure, (2) connected vehicle communication channels, and (3) in-vehicle software platform. It is the first CAV security testing platform supporting cross-component offensive and defensive testing services designed to capture the inter-relationships among the three key components. The project develops testing support by effectively combining techniques in optimization, statistical modeling, machine learning, network emulation, program analysis, and model checking, while supporting innovative use of known defense solutions, including trusted computing hardware, access control policy enforcement, and anomaly detection. The project implements and deploys the testing platform in real-world CAV testbeds and collaborates with industry partners for early adoption.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.