請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/71194
標題: | 交叉口右轉側撞之接受間距與交通衝突參數特性之研究 A Study of the Characteristics of Gap Acceptance and Traffic Conflict Indexes in Through with Right Collision at the Intersection |
作者: | Ying-Chieh Lin 林瑩潔 |
指導教授: | 許添本(Tien-Pen Hsu) |
關鍵字: | 交叉口安全,右轉側撞,後侵佔時間,接受間距,右轉決策,個體決策模式, Intersection Safety,Through and Right Collision,Post Encroachment Time,Gap Acceptance,Right-Turn Decision,Discrete Choice Model, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 側撞為台灣最常見肇事類型,占總肇事數量約四成,其中主要包含了右轉側撞和左轉穿越側撞,常見涉入車種為轉向車以及直行機車,對於機車駕駛傷害程度較高。過去常用肇事數量來衡量道路之安全,但可能面臨肇事資料不完整,輕微事件無紀錄,個人隱私問題,資料蒐集所需時間較長且肇事資料也代表事件已發生,近幾年多以衝突替代肇事判斷路口安全,能減少資料蒐集時間,也能在事件發生前做出主動式安全分析。
台灣因大量機車,擁有特殊混和車流形式,許多國外開發之智慧車輛系統,如自駕車、防撞系統等並不完全適用於台灣的車流情況,必須要先在地化。然而,現有關於右轉車流與同向直行機車流之接受間距之相關研究稀少。本研究最主要之目的乃是研究右轉汽車與同向直行機車駕駛人在都市內之交叉路口接受間距之特性。在號誌化路口,標線型快慢分隔路行下,右轉車有明顯的穿越行為,利用適當之研究方法,分析右轉汽車穿越直行機車接受間距之特性。且分為半間距(Lag)和間距(Gap),以及將接受(accepted)與拒絕(rejected)代表右轉車輛穿越與不穿越行為。 資料之調查選擇高雄與台中地區各一個交叉路口,利用錄影機進行資料之調查及整理,並以衝突與斷面位置關係衡量針對右轉側撞改善路口之前後差異,且使用適當之統計方法分析汽車右轉接受間距之特性。由分析結果顯示,近路口取消慢車道線並繪設分流指向線對於右轉側衝突是有效果的,但改善策略若無多方配合效果有限。對於右轉接受間距方面,可發現不同路口但路型相同之半間距(Lag)分布會相同,相同路口之半間距(Lag)與間距(Gap)分布也相同;以臨界間距來看臨界半間距較臨界間距大;本研究採用具有個體選擇觀念之二元羅吉特模式做為接受間距模式邏輯建構,並分成接受半間距模式與接受間距兩大模式,以右轉汽車與直行機車之車速與干擾區內機車數及(半)間距等影響變數,構建右轉汽車穿越直行機車接受間距決策邏輯。 Side collisions are the most common types of accidents in Taiwan, accounting for about 30% of the total number of accidents. The side collisions including mainly Through with Right Collision and Left Turn with Opposing Through Collision. Commonly involved types of vehicles are turning vehicle and straight scooter, which have a higher degree of damage to the scooter. In the past, the number of accidents was commonly used to measure the safety of roads. However, it may face incomplete information on accidents, such is minor incidents without records, personal privacy issues, long data collection time and the data also represents the accidents had happened. In recent years, the use of conflict alternatives to judge intersection safety, which can reduce the time required for data collection and also enable active safety analysis before the incident happened. Taiwan has a large number of scooters and has a specially mixed traffic flow. Many smart vehicle systems developed overseas, such as self-driving cars and collision avoidance systems, do not fully be useful in Taiwan and must first be localized. However, there are few studies on the correlation between the right-turn traffic flow and the straight-line traffic flow. The main purpose of this study is to study the characteristics of the gap acceptance between the right-turn car and straight scooters at an intersection in the city. In the environment, the signalized intersection, and the road type of the fast/slow lane with marking divider, and the right-turn car has obvious crossing behavior. Use appropriate research methods to determine the gap acceptance characteristics of the right-turn vehicle crossing through straight scooters. It is divided into Lag and Gap, and will accept and reject which represented the right-turn vehicles crossing and not crossing. The survey selected each intersection in Kaohsiung and Taichung city and used a video recorder to conduct surveys and collation of the data. And conflict with the positional relationship between the cross-sectional differences before the measure for improve intersections for through and right collision. And use appropriate statistical methods to analyze the characteristics of right-turn gap acceptance. According to the analysis results, the cancellation of the slow lane line near the intersection and the drawing of the diversion direction line can improve the through and right collision, but the improvement strategy has no effect if there is no multi-party coordination. For the right turn gap acceptance, we can find that at the different intersections with the same road type the Lag distribution will be the same, and the Lag and the Gap distribution at the same intersection will be the same. The critical Lag is longer than the critical Gap. In this study, using logit model as the gap acceptance decision model, which contains lag and gap models. Gap acceptance decision model is based the relevant variables such as speeds of the right-turn car and the straight scooter, number of scooters in interfering area and (lag)gap through traffic video analysis that influence gap acceptance behavior. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/71194 |
DOI: | 10.6342/NTU201801976 |
全文授權: | 有償授權 |
顯示於系所單位: | 土木工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-1.pdf 目前未授權公開取用 | 6.89 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。