Field Measurements,Modeling And Simulation Of Pedestrian-Vehicle Mixed Traffic Flows

Mixed traffic is a typical character in cities of developing countries.In road sections with large amount of pedestrians,such as in commercial streets and subway stations,there exist severe interferences between pedestrians and vehicles,which cause the inclination of transportation efficiency,and moreover the increase of traffic accidents.In this thesis,starting from field measurements,problems related to crossing behavior of pedestrians at signalized and unsignalized intersections and interactions of pedestrians and vehicles in narrow channels were investigated through constructing several improved cellular automaton models and conducting corresponding simulation and analysis,and a series of novel results was obtained.The main contents of the dissertation are as follows.(1)In order to understand the complex characteristics of interaction between pedestrians and vehicles,a set of field measurements of pedestrian crossing behavior were conducted at an unsignalized intersection in Wuzhou City,Guangxi Autonomous Region,China,to determine the decision-making behavior and velocity distribution of pedestrians in crossing such an intersection.It was found in the measurements that the decision-making distance is 10.8m and the critical safety distance is 15.3m at the unsignalized intersection.An empirical formula was obtained through the fitting of observed data.The observed data also show that the crossing speed of pedestrians at off-peak hour fluctuates around the mean speed and does not perfectly obey the normal distribution based on the D'Agostino test.The mean speed at off-peak hour is only 0.6 m/s when the waiting time is counted,whereas it is 0.8 m/s without counting the waiting time.(2)In order to describe clearly the interaction between pedestrians and vehicles on crosswalks at signalized intersections,an improved cellular automaton model was proposed for the mixed traffic flow under the influence of traffic signals.It was found that the interaction between pedestrians and vehicles are embodied at signal switching period because the motion of vehicles and pedestrians crossing behavior is separated by traffic light.There are two possible situations.One is that vehicles across the crosswalk do not have enough time to brake when the green light turns red.The other is that pedestrians have not yet crossed the crosswalk and hence will affect the moving of vehicles when the red light turns green.In the proposed model,the vehicle motion is described by a refined NaSch model and the deceleration zone is introduced.In the setting of crosswalk,a waiting area is introduced,in which those pedestrians without successfully crossing the street during the red-light period will remain.The interference between vehicles and pedestrians was studied respectively under periodic and open boundary conditions.In addition to studying the basic properties of the fundamental diagram for vehicle flow,we also studied the pedestrian waiting time.In particular,two characteristic quantities depicting the interaction between pedestrians and vehicles were defined,which have not appeared in the existing literature,namely,pedestrian-occupied vehicle travel time and vehicle-occupied pedestrian travel time.Through numerical simulation,the quantitative properties of vehicle flow-rate,pedestrian waiting time,pedestrian-occupied vehicle travel time and vehicle-occupied pedestrian travel time are given.The temporal and spatial evolution patterns of vehicle motion are discussed in detail.Finally,the phase diagram is given,which can fully reflect the qualitative character of the model.(3)The interaction between vehicles and pedestrians crossing crosswalks at unsignalized intersections was studied.In order to avoid traffic accidents,the alternating motion mechanism of pedestrians and vehicles is introduced,and the waiting time threshold is set respectively for the vehicles and pedestrians as two parts.One part can pass through the crosswalk when the waiting time exceeds its threshold,while the other part must wait.But the actual travel time of vehicles and pedestrians depends on the relative size of the number of vehicles crossing the street and the number of pedestrians.When the deceleration zone(waiting area)has no car(person),the pedestrian(vehicle)waiting for the street can begin to cross the street without waiting for the time to exceed its threshold.This model can well describe the competitive behavior of pedestrians and vehicles at unsignalized crosswalks.The interference between vehicles and pedestrians is studied respectively under periodic and open boundary conditions.In addition to studying the basic properties of the fundamental diagram for vehicle flow,we also investigated the physical quantities of pedestrians crossing characteristics and interaction between pedestrians and vehicles,such as pedestrian waiting time,pedestrian-occupied vehicle travel time and vehicle-occupied pedestrian travel time.The temporal and spatial evolution patterns of vehicle motion were discussed in detail.Finally,the phase diagram is given.Different from the situation of signalized crosswalks,there is a high traffic flow phase when the vehicle density or vehicle entry probability is large and the pedestrian arrival rate is small.(4)A cellular automaton model for coupled motion between pedestrians and a vehicle was constructed.The interaction between vehicles and pedestrians in a narrow channel was studied for the two cases that pedestrians move in the same direction and in opposite directions.The direction of the pedestrian movement is determined by the floor field,and the vehicle(and its affected area),regarded as a movable obstacle,causes the change of the floor field.Because of the continuous change of speed and the movement of the vehicle,the change of the floor field makes the direction of the pedestrians adjust accordingly.Through numerical simulation,the fundamental diagram of pedestrian flow under the typical parameters is obtained,and the two cases are compared.It was found that there is no significant difference in the fundamental diagram of pedestrians in the two cases when the density is very low and the density is very high.In these two cases,the distribution of pedestrians in the channel is more uniform.In between[p1,p2]density range,the two cases of pedestrian average speed and flow-rate are significantly different.This is due to the reverse movement of pedestrian and vehicle,the congestion ahead of the vehicle makes the average speed of the crowd significantly lower.We further researched the effect of vehicle maximum speed,pedestrian prejudgment time and vehicle width.It was found that the effect of vehicle width is the largest,and the time of pedestrian prejudgment is helpful to improve the speed of vehicle,while the effect of maximum speed of vehicle is relatively small.In the final part of this dissertation,the summary of the thesis and a prospect of further study of the mixed traffic flows are devoted.