Your browser does not support JavaScript!
智慧型信號處理實驗室(Intelligent Signal Processing)
不怕沒得學,就怕你不肯學,勇敢築夢!!
繁體

Recent

數據載入中...
智慧型運輸系統 – 先進駕駛輔助系統

昨夜準備與客戶開會資料幾乎整晚沒有睡好,臨時又找不到人可以駕駛,睡眼惺忪的你,勉強打起精神開著車。汽車開上高速公路,一路順暢沒車,很好開,可卻也像催眠曲般,讓濃濃睡意更加強烈。突然傳來警告聲,嚇得你睡意盡失,原來是汽車電腦系統發出的警告聲,提醒你未與前方出現的塞車車陣保持安全距離。「怎麼突然出現塞車車陣,還好有裝這警告系統」,冒著冷汗的你心想著。

前段故事中所提及的警告系統,稱為先進駕駛輔助系統(Advanced Driver Assistance System, ADAS)。ADAS和自動化駕駛不同,其功用在於輔助駕駛人而非完全取代駕駛人,預告可能發生的危險狀況,讓駕駛人提早採取因應措施避免交通意外。跟安全氣囊、防鎖死剎車系統(ABS)這類被動式安全系統不同,ADAS是一種主動式安全系統,可以在意外發生之前提醒駕駛人,駕駛過失是交通肇事主因,所以ADAS在車用電子產業有極高的需求,確實有其存在的必要性。ADAS是一個統稱,其實包含許多功能,包含:車道偏離警告、前車偵測及防撞警告、盲點偵測及警告、行人偵測及行為預測等。通常ADAS使用許多不同感測器,如雷射、超音波、全球衛星定位系統、攝影機等等,各有其優缺點。超音波感測器具有準確度高、成本低的優點,但是量測距離短;雷射感測器除同樣精準度高,測量距離較長,但是成本較高。因此,本實驗室採用CCD攝影機安裝於擋風玻璃上,除價格較低外還可獲得較多行車資訊,我們只需要撰寫適當演算法就能開發多種不同功能的ADAS。

本實驗室研究ADAS已經很多年,主要方向為車道偏離警告、前車偵測、前車測距及防撞警告。我們發展的車道偏離警告系統,由攝影機中擷取行車路況影像資訊,利用適當的影像前置處理強化道路線特徵,分離道路線與路面找出道路線的位置,設置影像畫面感興趣的區域(Region of Interesting, ROI),就可以藉由道路線與ROI位置判定汽車是否偏離車道,並對駕駛人發出適當的警告。前車偵測系統,可以偵測前車的位置,在ADAS中是非常重要的偵測功能,做法是攝影機擷取行車影像,經過影像處理擷取出物件特徵後,利用車輛的對稱性來尋找影像中前車的可能位置。行車影像中對稱性物件眾多,所以搭配車框比例與貝氏分類機器學習增加前車辨識正確率。前車測距系統是利用交通部規定道路虛線標線距離,當作影像中的距離與現實中的距離轉換依據。攝影機擷取的影像做完影像處理後,找出虛線道路線,使影像上的距離轉換成現實與前車的距離。進一步利用機器學習得到與前車距離的安全係數,判斷前車是否威脅到本車。

上述系統仍有許多可改善的地方,譬如:利用道路線並不會無故消失的特性將原已找到的道路線資訊,持續追蹤,降低系統運算量,提高整體的運算速度。本實驗室研究ADAS過程發展許多功能,現在已經將這些研究成果轉移到手機上,利用手機便於攜帶、容易架設且具有攝影鏡頭與高階處理器等特性,期望將此研究成果普及。

由於目前汽車尚未完成全自動化駕駛,先進駕駛輔助系統的存在對於汽車行駛安全有著一定的貢獻。除前述三項功能外,未來還有許多功能有待開發,將使得行車環境更加安全,像是追瞳功能可用來判斷駕駛目前是否分心或疲憊等,就讓我們拭目以待吧!

【延伸閱讀】

  1. Yu-Jie Jheng, Yu-Hua Yen, and Tsung-Ying Sun*, “A Symmetry-based Forward Vehicle Detection and Collision Warning System on Android Smartphone,” in Proc. International Conference on Consumer Electronics-Taiwan, May 26-28, 2015, Taipei, Taiwan.

  2. Yu-Hua Yen, Chih-Li Huo, and Tsung-Ying Sun*, “Adaptive Lane Departure Warning System on Android Smartphone,” in Proc. International Conference on Consumer Electronics-Taiwan, May 26-28, 2014, Taipei, Taiwan.

  3. Jiun-Hung Li, Chih-Li Huo, Yu-Hsiang Yu and Tsung-Ying Sun*, “Time Sequence based Lane-marking Identification,” in Proc. International Conference on System Science and Engineering, July 4-6, 2013, Budapest, Hungary.

  4. Chun-Wen Hung, Chih-Li Huo, Yu-Hsiang Yu and Tsung-Ying Sun*, “Road Area Detection based on Image Segmentation and Contour Feature,” in Proc. International Conference on System Science and Engineering, July 4-6, 2013, Budapest, Hungary.

  5. Fu-Hsaing Chi, Chih-Li Huo, Yu-Hsiang Yu and Tsung-Ying Sun*, “An Robust Tracking Approach for Vision-based Forward Vehicle Detection,” in Proc. International Conference on Fuzzy Theory and Its Application, pp. 352-359, Nov. 16-18, 2012, Taichung, Taiwan.

  6. Fu-Hsaing Chi, Chih-Li Huo, Yu-Hsiang Yu and Tsung-Ying Sun*, “An Effective Approach for Forward Vehicle Detection with Tracking,” in Proc. 1st IEEE Global Conference on Consumer Electronics, pp. 244-247, Oct. 2-5, Tokyo, Japan.

  7. Yu-Hsiang Yu, Fu-Hsaing Chi, Chih-Li Huo and Tsung-Ying Sun*, “A Self-Calibrating Distance Lookup Method For Camera-Equipped Vehicular System,” in Proc. 1st IEEE Global Conference on Consumer Electronics, pp. 384-387, Oct. 2-5, Tokyo, Japan.

  8. Chih-Li Huo, Yu-Hsiang Yu and Tsung-Ying Sun*, “Lane Departure Warning System based on Dynamic Vanishing Point Adjustment,” in Proc. 1st IEEE Global Conference on Consumer Electronics, pp. 440-443, Oct. 2-5, Tokyo, Japan.

  9. Fu-Hsaing Chi, Chih-Li Huo, Yu-Hsiang Yu and Tsung-Ying Sun*, “Forward Vehicle Detection System Based on Lane-Marking Tracking with Fuzzy Adjustable Vanishing Point Mechanism,” in Proc. 2012 IEEE International Conference on Fuzzy Systems within 2012 IEEE World Congress on Computational Intelligence, pp. 536-541, June 10-15, 2012, Brisbane, Australia.

  10. Chih-Li Huo, Yu-Hsiang Yu, Jhih-Cheng Syu and Tsung-Ying Sun*, “Vehicle Warning System for Land Departure and Collision Avoidance: Using Fuzzy Decision Making,” in Proc. 2011 IEEE International Conference on Fuzzy Systems, pp. 1554-1559, June 27-30, 2011, Taipei, Taiwan.

  11. Tsung-Ying Sun* and Wei-Chih Huang, “Embedded Lane-Marking Tracking System,” in Proc. 13th IEEE International Symposium on Consumer Electronics, pp. 627-631, May 25-28, 2009, Kyoto, Japan.

  12. Tsung-Ying Sun*, Wei-Chih Huang, and Shang-Jeng Tsai, “Robust Lane-Marking Tracking Algorithm based on Boundary detection and Kalman Filter,” in Proc. 9th International Symposium on Advanced Vehicle Control, Kobe, Japan, pp. 164-167, Oct. 6-9, 2008.

  13. Tsung-Ying Sun* and Shang-Jeng Tsai, “Fuzzy Adaptive Mechanism for Improving the efficiency and Precision of Vision-based Automatic Guided Vehicle Control,” in Proc. International Conference on Systems, Man and Cybernetics, Hawaii, USA, Oct. 10-12, 2005.

  14. Tsung-Ying Sun*, Shang-Jeng Tsai, Jiun-Yuan Tseng, and Yen-Chang Tseng, “The Study on Intelligent Vehicle Collision-Avoidance System with Vision Perception and Fuzzy Decision Making,” in Proc. IEEE Intelligent Vehicles Symposium, pp. 112-117, Las Vegas, Nevada, U.S.A., June 6-8, 2005.

  15. 鄭宇盛、孫宗瀛*, “應用於前車距離量測的道路虛線邊緣預測追蹤法,”2016 年民生電子研討會 (WCE 2016),國立東華大學,Nov. , 2016.

  16. 許智誠、游宇翔、孫宗瀛*,“基於快速道路線追蹤演算法與色彩模型轉換之道路車 輛檢測,” 第十八屆模糊理論及其應用研討會,東華大學,Dec. 3, 2010.

  17. 何吉哲、游宇翔、孫宗瀛*,“嵌入式圖形化汽車儀表板的設計與實踐,” 第十八屆模糊 理論及其應用研討會,東華大學,Dec. 3, 2010.

  18. Tsung-Ying Sun, Pei-Dan Chan, Shang-Jeng Tsai, Chan-Cheng Liu, and Wei-Chih Huang, “Face detection and eye tracking algorithm based on HSV Color Model for vehicle driver,” The 21th IPPR Conference on Computer Vision Graphics and Image Processing, CVGIP 2008, 佛光大學,Aug. 24-26, 2008.

  19. 孫宗瀛、黃承森、蔡尚錚,“粒子群最佳化演算法應用於車輛自動循跡系統,”2008年中華民國系統科學與工程會議 (NSSSE 2008),國立宜蘭大學,June 6-7, 2008.

  20. 孫宗瀛、黃偉秩、蔡尚錚,“以卡爾曼濾波器及鄰近峰值偵測法為基礎的道路線偵測演算法,”2008年中華民國系統科學與工程會議 (NSSSE 2008),國立宜蘭大學,June 6-7, 2008.

瀏覽數