Simulation Of Pedestrian Street Crowd Phenomenon Based On Agent Model

With the rapid advancement of China’s urbanization process,there are more and more crowded places,in which the population density is large,especially during some holidays.Once the crowded crowd is affected by emergencies during the movement,the pedestrian flow will gradually become chaotic,resulting in stampede accidents.Therefore,it is very important to study and analyze the movement phenomenon and potential safety hazards of people in dense places.For the research on the above problems,firstly,by analyzing several typical crowd stampede accidents,two influencing factors leading to crowd chaos are extracted,which are the channel blocking rate and crowd density caused by obstacles.According to the definition of crowded places,a pedestrian street is selected to observe the crowd movement phenomenon,which confirms the existence of these two influencing factors in the pedestrian street.Then,UWB high-precision positioning equipment,video equipment and timing equipment are used to carry out multiple groups of pedestrian kinematics parameter measurement experiments.In the first group of experiments,the average acceleration of pedestrians in the starting acceleration stage during normal walking is 1.05m/s 2、The stable average walking speed after acceleration is 1.46m/s and the average acceleration during emergency deceleration braking is-1.70m/s 2。 The second group of experiments measured that the average walking speed of dense people in pedestrian street was 0.70m/s.The third group of experiments measured the trajectory and speed changes of pedestrians when they kept normal walking speed,bypassed obstacles and avoided pedestrians in front,and obtained the linear function relationship between the obstacle avoidance distance of pedestrians avoiding obstacles in front and the side length of obstacles.The avoidance distance between pedestrians was 2.05 m.The agent model is constructed based on the idea of social force model,and the experimental results are used as parameters in the model simulation.Finally,the simulation scene of pedestrian street is established by using the constructed agent model to simulate and explore the macro movement phenomenon and micro speed change of pedestrian street crowd under the influence of obstacles.The results show that when the channel blocking rate caused by obstacles is less than50%,the impact on the crowd is relatively gentle,and when it is more than 50%,the impact on the crowd is more obvious.The impact of obstacles in the center of the channel on the crowd is greater than that on both sides of the channel.The simulation explores the macro movement phenomenon and micro speed change of pedestrian street crowd under the influence of crowd density,and processes the simulation results to obtain 0-3 people / m 2 The change curve of pedestrian speed within the range of crowd density,and the curve is compared with the curve studied by many scholars to verify the accuracy of the curve in this thesis.Simulate and explore the macro movement phenomenon and micro speed change of the affected population after taking common crowd guidance and control measures,and verify the effectiveness of the control measures from the macro phenomenon and micro speed levels.