Effectiveness of the pedestrian detection strategies and pavement markings: A driving simulator study

Abstract

Pedestrians are the most vulnerable group in the traffic and accounts for about 40,000 deaths each year worldwide. One of the main reasons is that drivers usually do not yield to pedestrian which creates serious conflicts between vehicles and pedestrians. The objective of this driving simulator study is to assess the effectiveness of detection based strategies and compare them with the untreated midblock marked crosswalk and pavement markings with pedestrian encircled. The detection based conditions include flashing LED lights placed before the crosswalk and animation based variable message sign VMS. Each condition was designed with a yield/stop controlled marked crosswalk. All the conditions were tested for two situations i.e. pedestrian present and pedestrian absent. Data from 62 drivers having a valid Qatari driving license were considered for analyses. The results showed that drivers yielding rate increased up to 98.3% in the detection based strategies (i.e. VMS and LED) compared to the pavement marking condition (i.e. 91%) and the untreated control condition (i.e. 88.7%). Regarding the vehicle-pedestrian conflicts, results revealed that serious and slight conflicts were significantly reduced in the VMS and LED conditions compared to the other conditions (i.e. Control and pavement markings). Furthermore, the detection based strategies were helpful in motivating drivers to reduce their traveling speed while approaching the pedestrian crosswalks. Based on the results from this study, the detection based strategies (i.e. LED and VMS conditions) are recommended for policy makers as a potentially effective treatments to improve safety at midblock yield/stop controlled crosswalks.