障碍物
行人
车头时距
社会力量模型
计算机科学
模拟
运输工程
工程类
地理
考古
作者
Han Xu,Jun Zhang,Weiguo Song,Yanghui Hu,Xudong Li,Xiangxia Ren,Longnan Yang,Hang Yu,Kechun Jiang
标识
DOI:10.1088/1742-5468/ac4c3f
摘要
Abstract One of the key problems for crowd management is to improve the evacuation efficiency of pedestrians. In this paper, we study pedestrian evacuation from a single exit room based on a social force model by considering the influence of a moving obstacle in front of the exit. The moving obstacle can improve evacuation efficiency for the pedestrians at different desired velocities. With a gap of 0.7 m to the exit and a speed of 0.7 m s −1 , it increases the evacuation efficiency by 29.94% to 149.66% for the pedestrians at the desired velocity between 0 m s −1 and 5 m s −1 . It can alleviate the ‘faster is slower’ effect and has a screening effect analogous to the effects of fixed obstacles. Besides, it reduces the mean and the relative standard deviation of time headway. The moving obstacle shows the more optimal evacuation results with the gap between the obstacle and the exit approximately equal to the integer multiple of the pedestrian diameter. The evacuation efficiency has a positive correlation with the speed of the moving obstacle where the speed is between 0.1 m s −1 and 0.5 m s −1 , while it is almost constant when the speed is between 0.5 m s −1 and 1.0 m s −1 . The moving obstacle promotes pedestrian flow for two reasons: the first reason is that the velocities toward the exit increase for most of the pedestrians, who are under the same crowd density as the scenarios without an obstacle. The second reason is that the crowd density near the exit is reduced, which benefits part of the pedestrians near the exit. The moving obstacle improves the order of the crowd motion leading to the velocities near the exit increasing. With a gap of 0.7 m to the exit and a speed of 0.5 m s −1 , it increases polarization by 34.1% to 80.7% for the pedestrians at the desired velocity between 0 m s −1 and 5 m s −1 .
科研通智能强力驱动
Strongly Powered by AbleSci AI