Alteration in delay time considering different pedestrian's crossing characteristics in the era of autonomous vehicles

Delay is one of the most crucial factors for both pedestrians and car drivers around pedestrian crossings. Drivers often do not yield to pedestrians, which may result in both delay and impatient pedestrian behaviour. This tendency may alter after introducing autonomous vehicles as the vehicles will follow the traffic rules in all cases. This study aims to estimate the delay time alteration at a simple zebra crossing using on-site measures and simulation. Roadside video recordings were carried out in Budapest, Hungary, to obtain the crossing decisions of pedestrian groups based on the approaching vehicle distance. We have determined the accepted vehicle distance vales for pedestrian groups that served as input data for microsimulation modelling. The novelty of the study is that the simulation involved autonomous vehicles that hold preset headways from the leading vehicle. The simulation was designed based on the traffic share of autonomous vehicles and the headways they kept. The main findings are that the travel time and stopping time for cars are higher if the modal share of autonomous vehicles is high. For pedestrians, however, we found a slight decrease in both travel times and stopping times. Moreover, we have proposed modifications to the simulation software (Vissim) to handle distance-dependent pedestrian decisions and drivers' failure to give priority. The results can be useful for road operators to estimate the road capacity in the era of autonomous vehicles and for software developers to formulate the simulated and real driving mechanism for autonomous vehicles.