Identifying the pedestrian position correctly is a challenging but crucial topic in the area of automatic driving. It is also an arising research focus that needs using the latest big data and data science techniques. In this paper, a hierarchical clustering (HCA) statistics learning algorithm has been applied to determine the location and amount of pedestrians detected by different vehicles. The vehicles have been equipped with a Pedestrian Autonomous Emergency Braking (PAEB) system. The inherent inaccuracy of the pedestrian sensing from these vehicles has been taken into consideration. It is found that the HCA method can generate robust results, since the proposed HCA structure also takes the vehicle ID information as additional block information between signals into the calculation. The HCA method determines the possible number of actual pedestrians by grouping the nearby pedestrians who are sensed and broadcasted by different vehicles. The simulation results have confirmed the effectiveness and applicability of the proposed HCA method. It is believed that the results using the HCA method can provide realistic information for vehicle PAEB systems to make better decisions to avoid crashing into pedestrians.