驾驶人纵侧向驾驶能力评价方法研究

为合理分配智能汽车人机协同共驾驾驶权，提高智能车辆的驾乘安全性和舒适性，本文提出驾驶人纵侧向驾驶能力的概念及其评价方法.对驾驶人的驾驶能力进行了定义和分析，并在此基础上设计了纵向跟车激励工况和侧向移动双移线激励工况，在搭建的驾驶人在环智能仿真平台上进行数据采集.建立了基于Hammerstein辨识过程的驾驶能力辨识模型，采用主成分分析法对驾驶能力辨识模型中的关键参数进行解耦和降维处理；通过客观蚁群聚类和主观量表分析相结合的分类方式，实现驾驶能力的分类；通过多元线性回归分析得到驾驶能力评价方程.结果表明，纵侧向驾驶能力辨识模型平均辨识及拟合精度均大于90%，经主成分分析及主客观分类处理后的纵侧向驾驶能力评价方程满足统计检验指标，具有良好的拟合及预测结果. 

Evaluation Method for Driving Capability in the Longitudinal and Lateral Scenarios

To allocate driving privilege in a reasonable way for shared control of intelligent vehicle, an evaluation method was proposed for driving capability in the longitudinal and lateral scenarios, to improve the safety and comfort of intelligent vehicles. Firstly, the driving capability was defined and analyzed and car-following stimulate for longitudinal scenario and moving double lane change stimulate for lateral scenario were designed. Data collection was conducted based on a driver-in-the-loop intelligent simulation platform (DILISP). And then, a driving capability identification model was established based on Hammerstein identification process and a principal component analysis (PCA) was used to decouple and reduce the dimension of the key parameters in Hammerstein model. Finally, a driving capability classification method was carried out based on a combination method of ant clustering algorithm (ACA) and subjective questionnaire. The evaluation equation for driving capability was developed by multiple linear regression(MLR). Results show that the proposed evaluation method for driving capability in the longitudinal and lateral scenarios can achieve accurate and reliable evaluation results. The mean identification accuracy of longitudinal and lateral driving capability identification model and fitting accuracy of evaluation equations can reach 90%.