臺灣博碩士論文加值系統

自動導航車是以攝影機取代人類的雙眼，以擷取路面資訊。本研究利用視差圖偵測自動導航車前進的路線與各種障礙物，再利用多項路面偵測器。辨別自動導航車是行駛於何種路面，最後再依據各項路面資訊判斷出自動導航車之最佳行駛路徑。我們於戶外沒有路線標示的道路上進行實驗，證明出我們的自動導航車仍具有良好的導航能力。

Similar to the function of human eyes, autonomous land vehicle (ALV) uses camera to acquire road information. In this paper, we adopt disparity map (DM) to detect ALV''s march path and various obstacles if may face to. Then we develop several road surface voters to recognize what kind of road surface the ALV drives on. Finally, according to the information we collected, the best navigation can be achieved. After experimenting with our ALV in an outdoor road without pavement markings, the proposed algorithms really work well.

摘 要 i
ABSTRACT ii
誌 謝 iii
TABLE OF CONTENTS iv
LIST OF FIGURES vii
Chapter 1 INTRODUCTION 1
1.1 RESEARCH MOTIVATION 1
1.2 SURVEY OF RELATED RESEARCHES 2
1.2.1 Road Segment Methods 2
1.2.2 Stereo Corresponding Methods 2
1.2.3 Obstacle Segment Method 3
1.2.4 Autonomous Vehicle Guidance Methods 3
1.3 THESIS ORGANIZATION 3
Chapter 2 ALV SYSTEM H/W ARCHITECTURE 4
2.1 HARDWARE ARCHITECTURE OVERVIEW 4
2.2 STEREO VISION 7
2.2.1 The Advantages of Stereo Vision 7
2.3 DESCRIPTION OF MOTOR CONTROL SYSTEM 8
Chapter 3 OBSTACLE DETECTION 9
3.1 NECESSARY INFORMATION OF PILOTING 9
3.2 THE MEANING OF DISPARITY MAP 9
3.3 OBSTACLE DETECTION WITH DM 13
3.4 PRODUCING EFFECTIVE BINARY DM 16
Chapter 4 ROAD SURFACE RECOGNITION 18
4.1 ON-LINE HYBRID OF DM AND MULTI-CLASSIFIER 19
4.2 ADVANTAGES OF THE HYBRID METHOD 24
4.2.1 The Capability to Pilot without Pavement Markings 24
4.2.2 Keep Old Efficacy of Original Voters 26
4.2.3 Fast Emergency Detection 27
4.3 SPEED UP FOR MULTI-CLASSIFIER 28
Chapter 5 ALV PILOT MODEL 29
5.1 EXCEPTED MOVEMENT 29
5.2 ADAPTIVE ADJUSTMENT OF ALV TURNING ANGLE 31
5.2.1 Hybrid of DM and the Best Voter of Road Surfaces 31
5.2.2 Scoring of Obstacle Base by Yellow Blocks of DM-line 34
5.2.3 Scoring of the Obstacle Body by Yellow Blocks 36
5.2.4 Scoring of Road by White Blocks 36
5.2.5 Scoring of the Other Blocks 37
5.2.6 Consider Distance of Obstacle for Angle-Score Table 38
5.2.7 Considering the ALV Safety Width for Avoiding Obstacle 41
5.3 Experimental Results 43
Chapter 6 CONCLUSION AND FURTHER RESEARCH 50
6.1 CONCLUSION 50
6.2 FURTHER RESEARCH 50
REFERENCES 51

[1]Massimo Bertozzi, Alberto Broggi, Alessandra Fascioli, “Vision-based intelligent vehicles: State of the art and perspective,” Robotics and Autonomous Systems, vol.32 (1), pp.1-16, Jun. 2000.
[2]Tian-Xiang Chen, Wei-Cheng Tu, and Rong-Chin, Lo, “A Study on Outdoor Guidance of Autonomous Land Vehicles Using Disparity Map Coordinated with Multi-Classifiers Based on BPNN,” The 21th IPPR Conference on Computer Vision, Graphics, and Image Processing, session D-4, no. 124, 2008.
[3]盧振育，基於立體電腦視覺及人工智慧策略搭配雷射光作室外自動車導航之研究，碩士論文，國立臺北科技大學自動化科技研究所，台北，2004。
[4] Yinghua He, Hong Wang, and Bo Zhang, “Color-Based Road Detection in Urban Traffic Scenes,” IEEE Trans. Intelligent Transportation Systems, vol. 5, no. 4, pp. 309-318, Dec. 2004.
[5]Jianbo Ma and Narendra Ahuja, “Region Correspondence by Global Configuration Matching and Progressive Delaunay Triangulation,” IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2, pp. 13-15, Jun. 2000.
[6]Il-Kyun Jung and Simon LACROIX, “A Robust Interest Points Matching Algorithm,” Eighth IEEE Conf. on Computer Vision, vol. 2, pp. 7-14, Jul. 2001.
[7]Stan Birchfield and Carlo Tomasi, ”Depth Discontinuities by Pixel-to-Pixel Stereo,” International Journal of Computer Vision, vol.35(3), pp. 269-293, Aug. 1999.
[8] Gary Bradski and Adrian Kaehler, Learning OpenCV: Computer Vision with the OpenCV Library, O’Reilly Press, 2008.
[9]Myron, Z. Brown, Darius Burschka, and Gregory D. Hager, “Advances in Computational Stereo,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 25, no. 8, pp. 993-1007, Aug. 2003.
[10]Liang Zhao and Charles E. Thorpe, “Stereo- and Neural Network-Based Pedestrian,” IEEE Trans. Intelligent Transportation Systems, vol. 1, no. 3, pp. 148-154, Sep. 2000.
[11]唐大崙，黃榮村，深度知覺與各類立體圖製作方法，1999.
[12] Nobuyuki Otsu, “A Tlreshold Selection Method from Gray-Level Histograms,” IEEE Trans. Systrems, Man, and Cybernetics, vol. 1 (1), pp. 62-66, Jan. 1979.
[13]Nicolas Hautière, Raphaël Labayrade, and Didier Aubert, “Real-Time Disparity Contrast Combination for Onboard Estimation of the Visibility Distance,” IEEE Trans. Intelligent Transportation Systems, vol. 7, no. 2, pp. 201-212, Jun. 2006.
[14]Yan Wang, Li Bai, and Michael Fairhurst, “Robust Road Modeling and Tracking Using Condensation,” IEEE Trans. Intelligent Transportation Systems, vol. 9, no. 4, pp. 570-579, Dec. 2008.