Research On The Synchronization Of Collective Dynamics And The Pedestrians Evacuation Of Panic System

Research of system science has a history for a lot of years. With the development of computer technology, the development of the theory of system science, especially the complex systems science was gradually perfect. Collective dynamics and pedestrian dynamics are two important problems of complex systems science. The collective motion is a ubiquitous phenomenon in the nature; collective dynamics is the study of the collective motion, it is based on the summary of nature biological cluster movement and uses the methods of mathematics and physics to analyze and forecast the movement of cluster. On one hand, collective dynamics can be used to explain the emergence of collective behavior in biological movement, on the other hand can also be applied to the artificial multiple individual system. Pedestrian dynamics problem is the collective dynamics problem in the human society. Compared to the animals, humans have a higher intelligence, the self-driven and the interaction among humans are much complex. The pedestrian dynamics can help us to ease the pressure of pedestrian traffic problems in the cities of growing population. The panic pedestrian evacuation is one of the important aspects of the pedestrian dynamics. Through perfecting the evacuation dynamics, we hope we can provide a better evacuation plan in an emergency.In this paper, we mainly study the two problems, one is the synchronization of collective dynamics; the other is panic pedestrians evacuation dynamics. The main content and innovation points are as follows:First, we reviewed the classical researches about collective dynamics in the past. Boid model, Vicsek model and Leader-follower model were three important models in the past researches; they simulated the cluster movement from different angles. Velocity synchronization in the cluster movement is an important phenomenon; we analyzed the factors that affected the synchronization in Vicsek model. To perfect the Vicsek model, we built a weighted model to make the model more close to the actual movement. In our new model, the influence of neighbors to center is associated with the angle between the line from their location to the center and the movement direction of the center individual:the angle is smaller, the influence is the greater. We set a weighted function containing a tunable parameter a and we found that there existed an optimal parameter αopt leading the system synchronization fastest. The optimal value is robust, αopt≌1 in different systems. This new model can achieve the collective motions synchronization faster, simulate the actual movement of creatures better and be more used in other areas because of the stability of the optimal tunable parameter.Pedestrian dynamics have been introduced before researching evacuation dynamics. In this paper, we have introduced the three classical methods in the pedestrian dynamics:hydromechanics, social force, and lattices-gas. In the research of panic pedestrian evacuation, most people used the microscopic models such as social force model and lattice gas model. The microscopic models not only can show the overall trend of motion also can reflect the individual movement and the interaction between pedestrians better. Most previous work studied the dynamics after the collective panic has been formed, and we have made a new discussion from the angle of the formation of collective panic. Based on the lattice-gas model, we classified the motion of pedestrians into two modes. The normal pedestrians can only choose the empty sites; the other panic pedestrians can choose the occupied sites and force the pedestrians in the sites to exchange position with them. We found that the panic pedestrians are lack of evacuation efficiency and cause more casualties. Further we think panic mood is infective and can make the normal pedestrians become panic. SIS model was used to simulate the panic spread process; we found that there is a fuzzy threshold of infection probability. The system can be in the "high-mix state" when the infection probability is greater than the fuzzy threshold. From these results, we think that the underlying mechanism of emergence of collective panic in pedestrian evacuations is similar to the process of rumor spread. In addition, we also found that, in the process of evacuation, whether pedestrian movement is fixed or not, the area that most wounded pedestrians appeared in is much the same, and this phenomenon is independent of the proportion of panic pedestrians and the infection probability.