高速公路事故發生時非事故車道上駕駛人感受與駕駛行為之研究

Abstract

事故發生常常造成車流非週期性的擁塞，此擁塞的原因除了事故車道阻塞，亦跟非事故車道上駕駛人的行為息息相關，但目前的文獻常常把事故影響區內非事故車道上的跟車行為視為一般正常情況的跟車，所以本文嘗試從駕駛者內心的觀點來解釋非事故車道上駕駛的跟車行為，並利用心理動量值描述駕駛者對週遭環境的感受，建構三個適用在高速公路事故發生時非事故車道上駕駛人駕駛跟車模式，1. 進入事故影響區的起始刺激、2. 未見事故時的跟車行為、3. 可見事故時的駕駛行為，並對建構出的模式進行參數校估與模擬，模擬結果發現，1. 當Phase 1心理動量門檻值為0.6時，即使沒有間距的限制，前後車亦不會相撞或造成過度剎車的情況。2. 模式確實可解釋駕駛者的專注或分心，且其對整體車流速度有明顯的影響。3. Phase 3確實可用來描述駕駛者通過事故時所發生膠頸行為的現象，因為膠頸行為，使得速度變的較為緩慢。4. 研究結果發現，此模式模擬下的駕駛者速度與車間距呈線性關係。

When car accidents occur, traffic often becomes quite congested around the site of the accident with many irregular driving patterns arising. In addition to the congestion on the lane where the accident occurs, the behavior of the drivers on the free flowing lane is closely linked to the slow speed of the through traffic. In this thesis the behavior of the drivers in the lanes other than the one in which the accident occurred will be analyzed through the drivers’ psychological status. In addition, this thesis will use a psychophysical momentum value to describe the drivers’ feeling toward the changed surroundings in order to develop these three car following models. These models can then be applied when car accidents happen on the superhighway. In plain form these models are: 1. The initial stimulus when entering the area where car accident has occurred. 2. The car drivers’ behaviors when they do not see the accident. 3 .Drivers’ behavior when they can see the accident. Afterwards, this thesis will collate and simulate parameters of those models. The simulated results are as follows: 1. When the threshold value of psychophysical momentum of the phase 1 is 0.6, even without the limited headway, the previous car and the following car won’t collide with each other nor brake overly hard. 2. The model explains drivers’ attention or distraction and the ways in which this would affect the speed of the flowing traffic. 3. Phase 3 explains the phenomenon of rubbernecking behavior, with which the traffic speed slows down. 4. The evidence from these models concludes that the drivers’ speed and headway are linear relationship with the simulation of this model.