人机共驾型智能汽车驾驶行为特性及人机交互方法研究

The intelligent vehicles are still far away from the level of full driving automation. The driving mode switching from automated to manual will take a long time. Meanwhile, the behavior of drivers is significantly different for intelligent vehicles when compared with traditional vehicles. This new feature of driver behavior will have an impact on driving safety and comfort. Therefore, human factor analysis, system identification and nonlinear modeling methods were used, simulated and field driving experiments were conduct in this study to address the following research topics: (1) the feature of driving workload for different driving mode switching; (2) the time-based variation feature of attention and alertness when driving in automated mode; (3) the evaluation methods for taking over ability when the switching from automated to manual; (4) the methods on cooperated hand over and optimization of human-machine interface. To achieve the abovementioned goals of this study, the key scientific questions of quantification parameters to measure workload, attention, and alertness need to be solved; the risk level modeling for takeover procedure and human machine interface mechanism of hand over need to be solved. In conclusion, the innovated driving behavior and switching mechanism for the intelligent vehicle could enhance the interdisciplinary of automotive engineering, artificial intelligence, and human factors. The findings of this study could support the theoretical design and driving of intelligent vehicles.

智能汽车在复杂交通环境中完全自动驾驶短期内难以实现，人驾/机驾混存状态将会长期存在。与传统汽车相比，人机共驾过程中的驾驶行为特性会发生很大的变化，而这一特性直接影响智能车辆行驶的安全性与舒适性。基于此，本项目综合运用人因工程、系统辨识与非线性建模等方法，利用实车实验、模拟实验等技术手段，系统研究（1）人机共驾过程中的驾驶负荷特性及变化规律 (2)智能驾驶状态下驾驶人反应力与警觉性时变规律 (3)人机切换过程中驾驶人接管能力评估研究 (4)人机共驾交互方法及协同切换等内容。重点突破人机共驾汽车驾驶人负荷与反应力及警觉性量化指标提取、人机切换接管过程风险态势建模及人机切换交互机制等关键科学问题，在人机共驾模式下驾驶行为特性与切换机制方面实现创新。项目研究成果可以促进车辆工程、人工智能与人因工程的学科交叉，为智能汽车在人机共驾阶段的安全行驶提供理论支持。

智能汽车在复杂交通环境中完全自动驾驶短期内难以实现，人驾/机驾混存状态将会长期存在。与传统汽车相比，人机共驾过程中的驾驶行为特性会发生很大的变化，而这一特性直接影响智能车辆行驶的安全性与舒适性。基于此，本项目综合运用人因工程、系统辨识与非线性建模等方法，利用实车实验、模拟实验等技术手段，系统研究（1）人机共驾过程中的驾驶负荷特性及变化规律 (2)智能驾驶状态下驾驶人反应力与警觉性时变规律 (3)人机切换过程中驾驶人接管能力评估研究 (4)人机共驾交互方法及协同切换等内容。重点突破人机共驾汽车驾驶人负荷与反应力及警觉性量化指标提取、人机切换接管过程风险态势建模及人机切换交互机制等关键科学问题，在人机共驾模式下驾驶行为特性与切换机制方面实现创新。项目研究成果可以促进车辆工程、人工智能与人因工程的学科交叉，为智能汽车在人机共驾阶段的安全行驶提供理论支持。