Pedestrian accident is a serious problem for the society. Pedestrian Collision Warning Systems (PCWS) are proposed to detect the presence of pedestrians and to warn the driver about the potential dangers. However, their interfaces associated with ambiguous alerts can distract drivers and create more dangers. On the other hand, Augmented Reality (AR) with Head-Up Display (HUD) interfaces have recently attracted the attention in the field of automotive research as they can maintain driver's gaze on the road. In this paper, we design a new PCWS with the AR cues and propose an experimental to evaluate the AR cues by assessing the driver's awareness of a pedestrian. At this stage, a fixed-based driving simulator is used for the study. Twenty five healthy middle-aged licensed drivers participate in the experiment. A car following task is proposed as the main driving task. Three levels of the driver's awareness of a pedestrian: the perception level, the vigilance level and the anticipation level are assessed through the observable outcomes. The results show that the proposed AR cues can enhance the driver's awareness of a pedestrian in all the three levels.