以卷積神經網路建構客運駕駛跟車距離之分類模式

Using Convolutional Neural Network to Construct Classification Models of Bus Drivers' Car-Following Distance

10.29416/JMS.202204_29(2).0001

盧宗成(Chung-Cheng Lu)；梁竣凱(Jyun-Kai Liang)；林玟妗(Wen-Chin Lin)

機器學習 ； 資料探勘 ； 卷積神經網路 ； 客運駕駛 ； 跟車行為 ； Machine Learning ； Data Mining ； Convolutional Neural Networks ； Bus Driver ； Car-Following Behavior

管理與系統

29卷2期（2022 / 04 / 01）

119 - 146

繁體中文

行車安全為政府所重視之公共運輸議題之一。若能了解客運駕駛行為之重要指標─駕駛跟車距離，業者便能夠有效地對駕駛員採取適當的行車管理措施。本研究之目的在以卷積神經網路建構一套跟車距離分類模型，將車載系統Mobileye所記錄之駕駛員跟車車距資料作為輸入資料，建立與訓練卷積神經網路分類模型，並藉由輸入之Mobileye車距資料來分類該駕駛員的跟車類型。本研究先將Mobileye所記錄之駕駛員跟車車距資料進行資料處理，並考量不同情境下之跟車距離，包括：天氣、車速及日夜三個因素，進一步設計七種不同的情境組合來篩選資料並產製車距矩陣圖，接著將車距矩陣圖輸入至卷積神經網路進行模型訓練及分類，利用訓練完成的模型來判斷所輸入之駕駛員車距圖片屬於何種跟車類型。實驗結果發現本研究所提出之分類模式在大多情境下皆有不錯的分類結果，其中以經圖片預處理之兩群分類準確率平均97%為最佳，顯示本研究之分類模型具備駕駛員跟車距離分類之能力，此模型可提供客運業者作為參考，藉由分類結果評估駕駛員的跟車距離類別，並針對不同類別駕駛員採行適當的管理措施。

Safety is one of the issues in public transportation that the government emphasizes. If bus companies are able to comprehend the most indicator of driving behavior of bus drivers-car-following distance, then companies can effectively adopt appropriate management measures for drivers. The purpose of this study is to develop classification models of bus drivers' car-following distance using convolutional neural networks (CNN). The headway data recorded by Mobileye systems were used as the input data to build and train the CNN classification models. The proposed classification models would facilitate determining the car-following types of bus drivers based on Mobileye headway data. We firstly processed the Mobile headway data and then considered different scenarios of car-following distance of drivers under three factors, including weather, speed and day-or-night. According to the three factors, we designed seven different scenarios and filtered the data by scenario. The filtered data were used to generate headway matrices for each driver. The matrices were the input data of the CNNs for training and classification. The trained models can be used to determine the car-following type of the input headway matrix of a driver. The experimental results show that the proposed CNN models have accurate prediction in most of the scenarios. Among them, the best classification results are the two groups that have been processed by the image preprocessor, and the accuracy rate is 97% on average. The results of this study can be used as a reference in classifying and determining the car-following distances of drivers. Bus companies can utilize the classification results to take actions on the driving behavior of drivers.

1. 今日新聞 NOWnews，「國道車禍釀 6 死 11 傷客運公司：會負起全部責任」，擷取日期：108 年 3 月 1 日（網頁：https://www.nownews.com/news/20170912/2605705/）。
2. 中央通訊社，「國道大客車與大貨車追撞 9 人受傷送醫」，擷取日期：108 年 3 月 1 日 （網頁：https://www.cna.com.tw/news/asoc/201805290059.aspx）。
3. Alpaydin, E.(2020).Introduction to machine learning.Cambridge:MIT press.
4. Bishop, C. M.(2006).Pattern recognition and machine learning.New York:Springer.
5. Cheng, Z.,Wang, W.,Lu, J.,Xing, X.(2020).Classifying the Traffic State of Urban Expressways: A Machine-learning Approach.Transportation Research Part A: Policy and Practice,137,411-428.
6. Choi, S.,Kim, J.,Kwak, D.,Angkititrakul, P.,Hansen, J. H.(2007).Analysis and Classification of Driver Behavior Using In-vehicle Can-bus Information.Biennial Workshop on DSP for In-vehicle and Mobile Systems
7. Chu, D.,Deng, Z.,He, Y.,Wu, C.,Sun, C.,Lu, Z.(2017).Curve Speed Model for Driver Assistance Based on Driving Style Classification.IET Intelligent Transport Systems,11(8),501-510.
8. Deng, C.,Wu, C.,Lyu, N.,Huang, Z.(2017).Driving Style Recognition Method Using Braking Characteristics Based on Hidden Markov Model.PloS one,12(8),e0182419.
9. Deng, L.,Yu, D.(2014).Deep Learning: Methods and Applications.Foundations and Trends in Signal Processing,7(3-4),197-387.
10. Duan, Y.,Lv, Y.,Wang(2016).Performance Evaluation of the Deep Learning Approach for Traffic Flow Prediction at Different Times.IEEE International Conference on Service Operations and Logistics, and Informatics
11. Feng, S.,Li, Z.,Ci, Y.,Zhang, G.(2016).Risk Factors Affecting Fatal Bus Accident Severity: Their Impact on Different Types of Bus Drivers.Accident Analysis and Prevention,86,29-39.
12. Fushiki, T.(2011).Estimation of Prediction Error by Using K-fold Cross-validation.Statistics and Computing,21(2),137-146.
13. Hao, K., "We Analyzed 16,625 Papers to Figure Out Where AI Is Headed Next," MIT Technology Review, 2019. (Available online at: https://www.technologyreview.com/2019/01/25/1436/weanalyzed-16625-papers-to-figure-out-where-ai-is-headed-next/)
14. LeCun, Y.,Bengio, Y.,Hinton, G.(2015).Deep Learning.Nature,521(7553),436.
15. Li, G.,Li, S. E.,Cheng, B.,Green, P.(2017).Estimation of Driving Style in Naturalistic Highway Traffic Using Maneuver Transition Probabilities.Transportation Research Part C: Emerging Technologies,74,113-125.
16. Lianzhen, W.,Yulong, P.(2014).The Impact of Continuous Driving Time and Rest Time on Commercial Drivers’ Driving Performance and Recovery.Journal of Safety Research,50,11-15.
17. Lovelace, R.,Fridman-Rojas, I.,Long, R.(2017).New Tools of the Trade? The Potential and Pitfalls of ‘Machine Learning’ and ‘DAGs’ to Model Origin-destination Data.GeoComputation 2017
18. Ma, X.,Dai, Z.,He, Z.,Ma, J.,Wang, Y.(2017).Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction.Sensors,17(4),818-834.
19. Pack, A. M.,Rodgman, E.,Cucchiara, A.,Dinges, D. F.,Schwad, C. W.(1995).Characteristics of Crashes Attributed to the Driver Having Fallen Asleep.Accident Analysis and Prevention,27,769-775.
20. Palat, B.,Saint Pierre, G.,Delhomme, P.(2019).Evaluating Individual Risk Proneness With Vehicle Dynamics and Self-report Data? Toward the Efficient Detection of At-risk Drivers.Accident Analysis & Prevention,123,140-149.
21. Perez, L.,Wang, J.(2017).,未出版
22. Polson, N. G.,Sokolov, V. O.(2017).Deep Learning for Short-term Traffic Flow Prediction.Transportation Research Part C: Emerging Technologies,79,1-17.
23. Russell, J., "Google's AlphaGo AI Wins Three-match Series Against the World's Best Go Player", TechCrunch, 2017. (Available online at: https://techcrunch.com/2017/05/24/alphago-beats-planetsbest-human-go-player-ke-jie/)
24. Shi, J.,Xiao, Y.,Tao, L.,Atchley, P.(2018).Factors Causing Aberrant Driving Behaviors: A Model of Problem Drivers in China.Journal of Transportation Safety and Security,10(4),288-302.
25. Sonnenstuhl, W. J.,Trice, H. M.(2018).Strategies for employee assistance programs: The crucial balance.New York:Cornell University Press.
26. Sutton, R. S.,Barto, A. G.(2018).Reinforcement learning: An introduction.Cambridge:MIT press.
27. Vujanić, M.,Antić, B.,Pešić, D.,Savićević, M.(2016).Testing the Psychophysical Characteristics of Professional Drivers-Can We Identify Unsafe Drivers?.Transportation Research Part F: Traffic Psychology and Behavior,42(1),104-116.
28. Yann, L., Corinna, C., and Christopher, J., The mnist database of handwritten digits. 1998 (Available online at: http://yhann.lecun.com/exdb/mnist)
29. 中華民國交通部統計處，交通統計月報，中華民國交通部，民國 107 年。
30. 江欣容,黃逸勤,范宜鴻(2019)。以大數據探討道路交通事故肇事原因。108 年道路交通安全與執法研討會論文集，桃園市:
31. 吳明隆(2018).SPSS 操作與應用－多變量分析實務.台北:五南書局.
32. 李宏毅(2016)。專題─人工智慧與 AlphaGo 什麼是深度學習。數理人文，10
33. 李佩憶(2017)。國立成功大學交通管理科學系。
34. 李嫺柔,陳慧婷,楊雅棠(2018)。從資源基礎理論觀點分析人力資本對績效影響之研究。全球管理與經濟，14(2)，107-121。
35. 張建彥,曾雅瑜(2006)。高速公路大客車跟車刺激－反應行為模式之建立。中華管理學報，7(3)，93-109。
36. 陳信丞(2014)。國立成功大學交通管理科學系。
37. 黃冠維(2018)。國立交通大學運輸與物流管理學系。
38. 葉政威(2017)。國立臺灣大學生物產業機電工程學研究所。