多危险源场景下的人群疏散仿真方法研究

发布时间：2019-01-05 09:45

【摘要】：随着经济社会的快速发展,人类的聚集性活动逐渐增多,人群运动中经常会发生一些难以预料的各种危险事件。这些危害的产生给人们的生命财产安全造成了极大的威胁,也严重影响社会的安全与稳定。深入分析研究危险发生后的人群运动状态,实现复杂场景下的人群疏散仿真已成为当前社会管理亟需解决的重要问题。针对上述问题,本文提出了一种多危险源下融入情绪感染机制的人群疏散仿真方法。该方法综合考虑人群应激运动过程中的心理状态与运动行为,使用多危险源下的情绪感染机制将人群心理状态实时反映到运动行为中。具体地,首先根据危险源的固有属性将其划分为不同的类型,针对这些不同类型的危险源引入危险场的概念,并定义了危险场内不同位置危险强度到个体恐慌情绪值的转化函数;随后,提出了一种多危险源下的个体情绪感染模型,用于描述恐慌情绪在人群中的传播过程;最后,通过在传统的RVO模型中融入情绪感染机制,提出了一种用于刻画人群运动行为的ERVO模型,该模型首次在人群仿真过程中实现了情绪影响与避障选择的结合。除情绪因素外,个体生理状态的改变也会影响人群的疏散过程,基于此,本文提出了一种多危险源场景下融合体力消耗的人群疏散方法。该方法在人群疏散过程中实时计算个体生理因素与情感状态之间的相关作用,得到了一种融入心理生理双重机制的人群疏散方法。具体地,首先对个体体力情况进行建模,得到个体运动过程中的实时体力消耗;随后,通过定量分析个体体力与情绪状态之间的关系,将体力消耗情况与情绪感染联系起来;最后,使用考虑体力消耗的ERVO模型模拟人群疏散过程。实验结果表明,本文提出的人群疏散方法可以在多危险源场景中生成较为真实的群体行为,并能充分反映个体恐慌情绪与体力情况的动态演变过程。综合心理状态与体力因素的共同作用,使得本文所提出的人群疏散方法更加符合实际情况,仿真结果也具有更强的真实性。
[Abstract]:With the rapid development of economy and society, the agglomeration of human beings increases gradually, and some unpredictable dangerous events often occur in the crowd movement. The emergence of these hazards to the safety of people's lives and property has caused a great threat, but also seriously affected the security and stability of society. It has become an important problem for the current social management to deeply analyze and study the movement state of the crowd after the occurrence of danger and realize the simulation of crowd evacuation in the complex scene. In order to solve the above problems, a method of crowd evacuation simulation is proposed, which is integrated into the mechanism of emotional infection under multiple risk sources. In this method, the psychological state and the exercise behavior of the crowd during the stress exercise were comprehensively considered, and the emotional infection mechanism under multiple risk sources was used to reflect the psychological state of the crowd to the exercise behavior in real time. Specifically, first of all, according to the inherent attributes of hazard sources, they are divided into different types, and the concept of hazard field is introduced to these different types of risk sources. And defined the transformation function between the danger intensity of different position and the value of individual panic emotion in the danger field. Then, a multi-risk individual emotional infection model is proposed to describe the spreading process of panic emotion in the population. Finally, by incorporating the mechanism of emotional infection into the traditional RVO model, a ERVO model is proposed to describe the behavior of crowd movement. The model combines emotional influence with obstacle avoidance in the process of crowd simulation for the first time. In addition to emotional factors, the change of individual physiological state will also affect the evacuation process of the crowd. Based on this, this paper proposes a crowd evacuation method combining physical exertion in multi-hazard scenarios. In the process of crowd evacuation, the correlation between individual physiological factors and emotional state is calculated in real time, and a method of crowd evacuation which is integrated with psychophysiological dual mechanism is obtained. Specifically, the physical performance of the individual is first modeled, and the real time physical exertion in the process of individual exercise is obtained, and then, by quantitative analysis of the relationship between the individual physical strength and the emotional state, the physical exertion is connected with the emotional infection. Finally, a ERVO model considering physical exertion was used to simulate the evacuation process. The experimental results show that the proposed evacuation method can generate more real group behavior in multi-hazard scenarios and can fully reflect the dynamic evolution of individual panic and physical conditions. The combination of psychological state and physical factors makes the proposed evacuation method more in line with the actual situation, and the simulation results are more authentic.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.9

【相似文献】