雾霾等恶劣气候对交通安全带来严重威胁。为分析雾霾天气下驾驶行为规律，利用高逼真度驾驶模拟器，构建了高速公路平直路段雾霾天气场景，开展了自由加速、紧急制动和跟驰试验；基于采集的试验数据利用遗传算法对IDM (Intelligent Driver Model) 跟驰模型进行了标定，并与晴天同路段标定参数进行了对比分析。结果显示，雾霾天气下IDM模型中起步加速度和舒适减速度参数并无显著差异，但安全距离、安全车头时距、期望速度等参数比晴天情况显著减小，由此表明雾霾天气下的交通冲突风险显著增大，应加强雾霾天气下的安全管理。

Adverse weathers like haze weather pose a serious threat to traffic safety. A high fidelity driving simulator was used to investigate the influence of haze weather condition on driving behavior. Scenarios with clear and haze weather conditions on straight section of a expressway were set up for the investigation, for which the weather conditions were defined by the visibilities. The three stages of free acceleration, emergent braking and car-following were conducted. Then, based on the experimental data, the genetic algorithm was proposed to calibrate the IDM (Intelligent Driver Model) car-following model. The results show that the haze weather condition has no significant influence on the start acceleration rate and the comfortable deceleration rate, but it significantly reduces the safety distance, the safety distance headway, the expected speed in IDM car-following model. This implies that the emergency traffic management should carefully pay attention to the significant increased potential safety conflicts in case of haze weather.