臺灣博碩士論文加值系統

根據研究調查指出高速交通碰撞事故所需的救援響應時間為影響道路交通事故死亡率的主要因素之一，雖然大多數的高速交通碰撞事故都能在事故發生後不久，由駕駛、乘客或熱心民眾協助報警得以獲得到相關的協助。然而，如果發生在郊區或夜間的高速交通碰撞事故，常因駕駛與乘車人員因劇烈撞擊而失去意識，在沒有人可以立即對外界發出求救訊息之狀況下，往往影響到交通碰撞事故傷者黃金救援時間，而造成重大身亡。因此，本研究實現一應用於行車高速碰撞事故即時報警之車聯網系統，主要包含碰撞感測模組、車載資通訊系統、與雲端資訊平台三大部分。當碰撞感測模組偵測到可能為發生高速交通碰撞事故時，車輛前方影像資訊會透過車載資通訊系統隨即上傳至雲端資訊平台，並藉由平台上之事故影像深度學習演算法來辨識該車輛前方影像是否為一碰撞事故，若為碰撞事故，雲端資訊平台可立即發出高速交通碰撞事故之報警訊息。經由實驗模擬驗證結果證明，本論文提出之系統具有極高之高速交通碰撞事故偵測準確率，因此，當高速交通碰撞事故發生時，除了可即時偵測外，亦可立即通報藉以縮短黃金救援時間進而有效提升傷者之存活率。

A research survey report pointed out that the rescue response time required for high-speed traffic accident collisions was one of the main factors affecting the road traffic accident mortality rate. Most traffic collision accidents could be got relevant assistances by drivers, passengers, or kindhearted persons. However, if a traffic accident occurs at night or in the suburbs, often driving or occupants lose consciousness due to violent impact, no one can immediately send a distress message to the outside world and affect the golden window the traffic collision accident.
Therefore, this research implementation of an Internet of vehicle system for high-speed traffic collision accident detection with an emergency alert mechanism, mainly including a collision sensing module, an in-vehicle infotainment (IVI) platform, and cloud information platform. When possible traffic collision event is detected, the vehicle front image information will be uploaded to the cloud information platform through the IVI platform, and the accident image depth learning algorithm on the platform will be used to identify whether the image in front of the vehicle is a collision accident. If it is a collision accident, the cloud information platform can immediately issue an alarm message for a high-speed traffic collision accident. The experimental simulation results prove that the system proposed in this paper has extremely high detection accuracy of high-speed traffic collision accident. Therefore, when a high-speed traffic collision accident occurs, in addition to immediate detection, it can be immediately notified to shorten the gold rescue. Time will effectively improve the survival rate of the injured.

摘要 V
Abstract VI
致謝 VII
目次 VIII
表目錄 X
圖目錄 XI
第一章 緒論 1
1.1 前言 1
1.2 研究動機與目的 1
1.3 研究方法與步驟 2
1.4 論文架構 4
第二章 參考文獻 5
2.1 基於異質網路結構下的交通事故感測 5
2.1.1異構網路高速公路的事故感知與報警方法 5
2.1.2 智能交通錐之事故檢測與智慧交通識別 8
2.2 基於車聯網之交通事故感測 10
2.2.1 應用於智慧型手機之碰撞事故檢測 10
2.2.2 eCall 12
2.3 文獻探討結論 14
第三章 系統實現 15
3.1 系統架構 15
3.2 感測層: 碰撞事故感測器模組設計 16
3.2.1 碰撞事故感測器選用 17
3.2.2 碰撞事故感測器通訊方式 18
3.3 網路層: 車載資通訊系統 19
3.3.1車載資通訊系統平台選用 20
3.3.2 車載資通訊系統UI設計 21
3.3.3 車載資通訊系統架構 23
3.3.4 偵測碰撞事故之門檻值設定 27
3.4 應用層: 雲端資訊平台實現 29
3.4.1 雲端資訊平台設計 30
3.4.2 雲端資訊平台UI設計 31
3.4.3 雲端資訊平台之事故影像深度學習預測實現 33
3.5 系統應用情境 34
第四章 實驗結果與討論 35
4.1 實驗環境設置 35
4.1.1 實驗室 35
4.1.2 實體車輛 37
4.2 實驗目的 37
4.3 實驗流程 38
4.4 實驗結果 40
4.4.1 車載資通訊整合異質網路結果 40
4.4.2 雲端深度學習之結果 43
4.4.3 應用於行車高速碰撞事故即時報警系統整體實驗 45
4.5 實驗結果討論 47
第五章 結論與未來展望 48
5.1 結論 48
5.2 未來展望 49
參考文獻 50

[1]“Connected Car Market - Global Industry Analysis, Size, Share, Growth, Trends and Forecast 2013 - 2019,” Transparency Maket Research, pp. 97, 2013.
[2]L. Roches. Powertrain, Body, Chassis and Safety by Strategy Analytics. Accesed: Mar. 2018. [Online]. Availabe:
https://www.strategyanalytics.com/access-services/automotive/powertrain-body-chassis-and-safety/market-data/report-detail/presenter---automotive-electronics-system-demand---april-2016?Related#.WzBOh1UzaUk
[3]World Health Organization (WHO), Global Status Report on Road Safety 2015, 2015. Accessed: Mar. 2018. [Online]. Availabe:
http://www.who.int/violence_injury_prevention/road_safety_status/2015/en/
[4]A. Kumar, S. Butt, O. Abudayyeh, T. Fredericks, and M. Kuk, “A Statistical Model for Emergency Medical Services Response to Motor Vehicle Crashes in Michigan,” Research and Creative Activities Poster Day, 238, 2017.
[5]European Commission, Speed and the Injury Risk for Different Speed Levels. ccesed: Mar. 2018. [Online]. Availabe:
https://ec.europa.eu/transport/road_safety/specialist/knowledge/speed/speed_is_a_central_issue_in_road_safety/speed_and_the_injury_risk_for_different_speed_levels_en
[6]姜海洋，李建兵，葉珊珊，徐靜波，2017，<異構網路高速公路事故感知與報警方法>，《Microcontrollers & Embedded Systems》， 7，頁34–38 。
[7]S. Kantawong and T. Phanprasit, "Intelligent traffic cone based on vehicle accident detection and identification using image compression analysis and RFID system," in Proc. of the 2010 ECTI International Confernce on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON2010), Chiang Mai Thailand, 2010.
[8]H. Sharma, R. K. Reddy, and A. Karthik, “S-CarCrash: Real-Time Crash Detection Analysis and Emergency Alert Using Smartphone,” in Proc. 2016 IEEE International Conference on Connected Vehicles and Expo (ICCVE’16), pp. 36-42, 2016.
[9]M. Hamid, Ali, S. Zainad, and Alwan, “Car Accident Detection and Notification System Using Smartphone,”IJCSMC, 4, pp. 620-635, April 2015.
[10]S. Howard, L. Kuehne, C. Thomas,and O. Kevin Vincent, “Preliminary Statement of Policy,” NATIONAL HIGHWAY TRAFFIC SAFETY ADMINISTRATION, 2016.
[11]D. Dumitrescu and R. Ropot, “eCall,” ITS Romainia. Accessed: Mar. 2018. [Online]. Availabe: http://its-romania.ro/en/h_ecall.php
[12]黃世霖，張金換，王曉冬， 2000，<汽車碰撞與安全>。取自https://books.google.com.tw/books/about/%E6%B1%BD%E8%BD%A6%E7%A2%B0%E6%92%9E%E4%B8%8E%E5%AE%89%E5%85%A8.html?id=jsIqd1HPRtYC&redir_esc=y
[13]黃興閎，2006，<感測器於實車碰撞測試之應用車輛研測資訊>，頁8 - 17。
[14]Jules White, Chris Thompson, Hamilton Turner, Brian Dougherty, Douglas C. Schmidt,”WreckWatch: Automatic Traffic Accident Detection and Notification with Smartphones,” Mobile Networks and Applications, Vol.16 No. 3, p.285-303, June 2011.