道路安全決策支援系統之研究

The Research of Safety-Road Decision Support System

孔祥聖(Sam Kong)

指導教授 ： 姚修慎

元智大學/資訊學院/資訊工程學系/碩士(2007年)

摘要

決策支援系統是個人性化的界面，它能夠有效地提示使用者，並且，在最短時間內將所想要的解答找出。現今還是有不少地方依然隱藏著不少的危機，而危險的發生多來自於危險的駕車行為，今我們試著發展出一個以行車記錄器基礎的應用，首先我們試著要建立一些有效的危險行為專用指標，利用指標的方法有效定義出危險路段，並且，藉由系統的能力將危險路段的影片播放出來，提供決策者作出正確的決定，盡而達到降低交通意外的危險率。

我們所使用的數位行車記錄器，包含了不少的資料，主要包含了行車中的三個加速度值(前後加速度、左右加速度以及方向角加速度)、GPS資料等等，記錄了行車中的過程資料，資料的記錄存放於車機內的記憶體，並且，可以很容易地將資料轉移至電腦中，使得電腦可作更進一步的分析。而數據的分析更是我們發展系統的重要部分，透過數據的分析，使得我們可以定義各種有可能的危險行為。

關鍵字

行車記錄器,車機,決策支援

並列摘要

Decision Support System is human-based interface. It can give some some knowledge for you, and finds the answer what you want quickly. There are still exists some traffic accidents. Usually, the accident comes from hazardous human-driving. We try to develop a safety-recorder-based system, and we will use some special method to detect the danger pattern. With our method, we can easily present the dangerous road by our designed dangerous index. Then, our system can show the video for the policy-making man. Finally, occur rate of the traffic accident can be reduced.

Our safety-recorder contains many data. The most important data include forth and back acceleration, side acceleration, direction acceleration and GPS data etc. All the data will be recorded when the car's power is turn on, and the data will be recording into the memory stick. The memory stick's data can easily be read to computer by card reader, and the computer can translate the binary data to text data for use. The data analysis is one of our most important parts. With the result, we can define the danger pattern finally.

並列關鍵字

Safety-recorder,SRcomm,Decision Support System,DSS

參考文獻

[1] Kuhler, M. and D. Karstens, 1978, “Improved driving cycle for testing automotive exhaust emissions, ” SAE Technical Paper Series 780650.

[2] Ericsson, E., 2001, ”Independent driving pattern factors and their influence on fuel-use and exhaust emission factors, ” Transportation Research Part D, Vol. 6, pp. 325-345.

[4]. Flahaut, B. (2004). Impact of infrastructure and local environment on road unsafety. Logistic modeling with spatial autocorrelation. Crash Analysis and Prevention, 36(6), 1055– 1066.

[5] Flahaut, B., Mouchart, M., San Martin, E., & Thomas, I. (2003). The local spatial autocorrelation and the kernel method for identifying blackzones. A comparative approach. Crash Analysis and Prevention, 35(6),991– 1004.

[8] Identification of Human Driver Models in Car Following，IEEE，Burnham et al.，1974