The Behavioral Analysis and Modeling of Lane Changes in Traffic Micro-simulation

摘要

In traffic system microscopic simulation, the processes of vehicle lane changes need to be described with high fidelity, and this is called lane change modeling. However, there are two defects in the present lane change models: one is the lack of researches on the attainable area of Normal Lane Change (NLC), the other is the ignorance of the phase during which the front wheel turning angle is 0 degree in the process of a lane change. Therefore, this paper discusses the attainable area of NLC, which provides rules to determine if vehicle reversing is necessary to perform a lane change successfully. And by analyzing drivers' behavioral characteristics at NLC, the traditional two-phase and three-phase lane change models are improved in this paper. The process of a lane change is divided into four phases: turning-angle phase, approaching phase, counter-turning-angle phase, and adjustment phase. The rationality of the model is testified by analyzing the actual images of lane change taken on the streets of Xi'an, China.