L3级自动驾驶技术要求驾驶权可以在驾驶员（人）与自动驾驶系统（机）之间切换。安全的人机交互机制是其核心关键技术，而准确地量化识别驾驶意图是人机交互实现的基础。实际驾驶过程中，驾驶员的行为在瞬时具有随机性与不确定性，但在长时间跨度内具有统计规律。如何利用统计学工具深入挖掘驾驶员行为的特征与规律是量化识别驾驶意图的关键，也是目前尚未解决的难题。线控转向（SBW）系统是自动驾驶的重要执行器，也是人机交互的关键部件。因此，本项目提出了人机共融线控转向系统，利用方向盘力矩来量化识别驾驶意图，从而建立安全的人机交互机制。项目主要内容包括：1）设计满足ASIL-D级功能安全的SBW系统方案；2）研究驾驶意图量化识别方法和主动驾驶权切换方法；3）研究驾驶风险认知评估方法和被动驾驶权的切换方法；4）搭建SBW测试平台，形成迭代式研究过程。研究工作将为下一代SBW系统的研发与L3级自动驾驶技术提供理论支撑。

L3 level automatic driving technology requires that the driving control right be switched between the driver (human) and the automatic driving system (machine). Safe human-machine interaction mechanism is the key technology, and accurate recognition of driving intention is the basis of human-machine interaction. In the actual driving process, the driver's behavior is stochastic and uncertain in the instantaneous, but has statistical law in the long-time span. How to use statistical tools to mine the characteristics and laws of driver behavior deeply is important to identify driving intention quantitatively, but it is also an unsolved problem. Steering by wire (SBW) system is an important actuator of automatic driving, and a key part of human-machine interaction. Therefore, this project puts forward a human-machine-cooperative SBW system, which uses the steering wheel torque to quantify the driving intention, so as to establish a safe man-machine interaction mechanism. The main contents include: 1) a SBW system scheme meeting ASIL-D level functional safety will be designed; 2) methods of driving intention quantitative identification and active driving control right switching will be stutied; 3) methods of driving risk cognitive assessment and passive driving right switching will be studied; 4) SBW test platform will be built to realize iterative research process. The research work will provide theoretical support for the research and development of the next generation SBW system and L3 level automatic driving technology.