Research On Crowd Evacuation Path Planning Method And Simulation Based On Multi-source Information

With the development of social economy and science and technology,modern buildings have become larger,more complex in structure,and more powerful in functions.Large public buildings such as shopping plazas,office buildings,medical and scientific education centers have increased dramatically.Large public buildings are often densely populated,with different internal structures and diverse functions.Even if some places have floor plans,there are still some people who cannot find a place because they are lost.In this case,if a sudden disaster such as a fire,earthquake,gas leak,etc.,due to people's panic during the disaster and the lack of effective evacuation guidance,the evacuation of people in the building will be inefficient.Therefore,an efficient crowd evacuation planning system has gradually become the focus of research.When preventing disasters,simulating crowd evacuation scenarios,and formulating evacuation plans in advance,the traditional field drill method has many restricted conditions,mobilizing a large number of personnel is not easy to operate,and wastes power,material resources and time,and the effect is not good.Computer simulation is different from field exercises,which can effectively solve these problems,it is convenient,flexible and saves time.However,the more realistic simulation method consumes too many system resources,and the method of reducing the system resource occupancy rate is often not realistic enough.The simulation method is needed to show the scene information and ensure the lower system resource consumption.Aiming at the problems of crowd evacuation path planning and simulation,the main research contents of this thesis include:(1)A crowd evacuation path planning system integrating multi-source information is proposed.At present,there are more and more complex buildings,and efficient indoor evacuation planning needs to fully understand the structure of the building,the number of indoor people,the location of the disaster,the scope of the disaster,and the available roads,and then formulate a reasonable plan according to the real-time situation.Only using information from a single source cannot fully grasp the situation of the entire scene.In order to improve the efficiency of crowd evacuation in multi-storey buildings,we make full use of various information in the evacuation scene and integrate the multi-source information obtained from the Internet of Things,Bluetooth devices,web cameras and mobile phones to formulate a reasonable evacuation plan.(2)A crowd evacuation path planning method based on swarm intelligence algorithm is proposed.Due to the large amount of calculation data and low calculation efficiency of the traditional model,we propose an improved artificial bee colony algorithm,and apply this method to multi-floor crowd evacuation path planning.Based on the original artificial bee colony algorithm,we optimize the initial state of the food sources and improve the objective function to adapt to the crowd evacuation of multi-story buildings.Experimental results show that this method effectively improves the calculation efficiency,and can plan the path in real time to achieve the purpose of rapid evacuation.(3)A crowd evacuation simulation method combining multi-source information is proposed.In order to improve the authenticity of crowd evacuation simulation,a method based on image segmentation is proposed to improve the diversity of crowd model and build a crowd simulation character model.Aiming at the problems of complicated traditional scene modeling and low evacuation simulation efficiency,a method based on 3D scene map optimization is proposed.By constructing the real scene,the 3D scene model scaled to the real scene is completed.The scene optimization problems that appear in the simulation scene based on the real scene are optimized from obstacles,falling crowds,and a single room.Finally,we use web cameras,Bluetooth,mobile phones and other devices to collect pedestrian movement trajectory,number of people and coordinates and other data,and then import these real evacuation data into the simulation to improve the authenticity of the evacuation simulation.This method combines information from multiple sources,which not only guarantees the authenticity of the simulation,but also optimizes the crowd evacuation simulation,reduces the amount of calculation,improves the calculation efficiency,and lays the foundation for the real-time planning of crowd evacuation planning.Finally,the above improved methods are applied to crowd evacuation path planning and simulation.In consideration of the evacuation path length,evacuation time,safety and other factors,the evacuation simulation research and analysis of the school teaching experimental building are carried out.The experimental results prove that the method proposed in this thesis can effectively carry out crowd evacuation planning and real simulation,and has high practicality.