2020-02-012024-05-17https://scholars.lib.ntu.edu.tw/handle/123456789/690889摘要:近年來車聯網與自駕車的技術蓬勃發展,但是相對於傳統的車內通訊與車輛控制,車聯網的無線通訊與自駕車的機器學習,即使是其模型本身,皆無法對於系統提供保證。除此之外,車聯網的無線通訊與自駕車的機器學習也都可能受到惡意攻擊,進而影響整體系統與人員的安全(safety)。針對這些問題,此計畫考量車聯網與自駕車幾項代表性的應用,例如智慧型路口管理、合作適應巡航控制、軟體空中更新、動態地圖生成,進而研發適用於車聯網與自駕車之金鑰管理系統、共識維持演算法、入侵檢測系統。此計畫的目標為在存在通訊錯誤與惡意攻擊的情況下,對車聯網與自駕車提供系統穩健性與系統安全性(security),進而對系統功能正確性提供保證。除此之外,由於系統有限的運算與通訊資源與嚴格的系統限制,此計畫也強調整體系統整合,以上針對系統穩健性與系統安全性的解答必須符合系統的設計要求。我們預期此計畫,在系統穩健性與系統安全性可能成為車聯網與自駕車是否能夠成真的關鍵時,能對於車聯網與自駕車之系統穩健性與系統安全性提供創新性的解答。<br> Abstract: The technology of connected vehicles (Internet of Vehicles) and autonomous vehicles is blooming in recent years. However, the vehicle-to-X wireless communication may suffer network faults due to the open and uncontrolled communication environment, and the machine learning mechanisms for autonomous driving cannot provide guarantees, especially when the corresponding training data is missing. Furthermore, the wireless communication and the machine learning mechanisms are vulnerable to malicious security attacks which can lead to catastrophic incidents. To address these robustness and security concerns in the system level, we plan to consider representative applications of connected and autonomous vehicles, such as intelligent intersection management, cooperative adaptive cruise control, over-the-air update, dynamic map generation, etc. We propose to develop key management systems, consensus algorithms, and intrusion detection systems applicable to connected and autonomous vehicles and guarantee the functional correctness, even with the existences of faults, errors, and security attacks. Besides, we intend to emphasize the system integration for the proposed solutions. This is because the resource, such as computational capability and communication bandwidth, on a connected and autonomous vehicle is usually limited, and there are tight and hard real-time deadlines. Any solution for robustness or security must be compatible with existing systems without violating system requirements. We expect the project to lead the research of robustness and security of connected and autonomous vehicles, especially when robustness and security may be a bottleneck of the realization of connected and autonomous vehicles in the future.車聯網自駕車系統安全性(Security)系統穩健性金鑰管理系 統共識維持演算法入侵檢測系統系統整合。Connected VehiclesAutonomous VehiclesSystem-Level SecuritySystem-Level RobustnessKey Management SystemConsensus AlgorithmsIntrusion Detection SystemSystem Integration.車聯網與自駕車之系統安全性與系統穩健性(2/5)