The Research On The Overtaking Behaviour Of Pedestrian Social Groups Based On Social Forces

With the continuous expansion of urban scale and the rapid expansion of the urban population, city traffic problem is becoming more and more serious. As an important part of city traffic, pedestrian traffic influence is more and more prominent. It is often seen that pedestrian jams take place in public places such as subway pedestrian passage, overpasses, large venues, sidewalks and other places. Then it puts forward new requirements for pedestrian imitation in public facilitiesPedestrian simulation model can be divided by differences in scale macro models, the medium model and the microscopic model. The current study is mainly aimed at the macro and micro aspects of conduct, while the medium level-peer group and its effects on macroscopic traffic flow is less. Social force model which based on molecular kinetic theory is pedestrian dynamics model that can accurately describe the relationship between individuals and reproduce the phenomenon of self-organization, but the overtaking behavior of a single pedestrian is inconsistent with the reality of the situation in the traditional social forces model. The group will be more obvious to follow the phenomenon under existing group model based on social forces, producing a relatively long lag and it’s not realistic.This paper summarizes the pedestrians (group) walking characteristics, analyzes the traditional social force model and building process and principles of Moussaid group model which laid a theoretical foundation for constructing group models based on social forces in this paper.The main research work are as follows:(1)Analysis the principles of traditional social force model which is based on molecular kinetic theory and we proposed improving direction to the defects, Then, we build an improved social force model. This model introduces the concept of visual areas and pedestrian decisions, amended the desired direction pedestrian and recreates the overtaking scene under low density.(2) Elaborate the building process and principles of Moussaid group model, which includes the visual force, attraction force and repulsion force. analyzed defects of Moussaid model when passing pedestrians, amended traveling desired position of the group based on the overtaking behavior of group pedestrians. Then we establish an improved model of groups,which can let the group traveling smarter and more realistic.(3) Using MATLAB programming, building simulation environment within the subway tunnels. We compared and analyzed peer group behavior under different scenarios, verifying the validity of the model by comparison of the data display and analysis scenarios.