Page 21 - Autonomous Mobile Robots

1 Visual Guidance for

Autonomous Vehicles:

Capability and

Challenges

Andrew Shacklock, Jian Xu, and Han Wang

CONTENTS

1.1 Introduction ............................................................ 6
1.1.1 Context......................................................... 6
1.1.2 Classes of UGV ............................................... 7
1.2 Visual Sensing Technology ........................................... 8
1.2.1 Visual Sensors ................................................. 8
1.2.1.1 Passive imaging .................................... 9
1.2.1.2 Active sensors ...................................... 10
1.2.2 Modeling of Image Formation and Calibration .............. 12
1.2.2.1 The ideal pinhole model ........................... 12
1.2.2.2 Calibration .......................................... 13
1.3 Visual Guidance Systems ............................................. 15
1.3.1 Architecture ................................................... 15
1.3.2 World Model Representation ................................. 15
1.3.3 Physical Limitations .......................................... 17
1.3.4 Road and Vehicle Following .................................. 19
1.3.4.1 State-of-the-art ..................................... 19
1.3.4.2 A road camera model .............................. 21
1.3.5 Obstacle Detection ............................................ 23
1.3.5.1 Obstacle detection using range data ............... 23
1.3.5.2 Stereo vision ........................................ 24
1.3.5.3 Application examples .............................. 26
1.3.6 Sensor Fusion ................................................. 28
1.4 Challenges and Solutions.............................................. 33
1.4.1 Terrain Classification.......................................... 33

5

© 2006 by Taylor & Francis Group, LLC

FRANKL: “dk6033_c001” — 2006/3/31 — 16:42 — page5—#5