본문 바로가기 주메뉴 바로가기
22대 대한민국 국회 국회정보길라잡이 국회의원검색
회원가입 로그인
HOME

정보검색

소장정보 검색

국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

고급검색

인명/단체명

저자정보 상세정보
인명/단체명을 입력하세요.

키워드

  • 대표어

  • 외국어

-

목차보기

Title Page

Abstract

Contents

1. Introduction 13

1.1. Motivation 13

2. Background 23

2.1. General Methods for Lane Detection 23

2.2. Light Intensity Detection for Dual Light Sensor 28

3. Design of Pedestrian Target Selection with Funnel Map 31

3.1. Introduction of Pedestrian AEB System 31

3.1.1. AEB System Architecture 31

3.1.2. AEB System Control Concept 35

3.1.3. Configuration of pedestrian target selection 36

3.2. Proposed Pedestrian target selection 38

3.2.1. Candidate Target Selection Based on a Funnel Map 38

3.2.2. Calculation of the Prediction Time 40

3.2.3. Calculation of the Predicted Path of a Funnel Map 42

3.2.3. Distance to collision Calculation 46

3.3. Experimental Results 51

3.3.1. Test Environment for Pedestrian AEB 51

3.3.2. Test Scenarios 53

3.3.3. Experimental Results 56

3.3.4. Real Road Tests 72

3.3.5. Limitations 76

4. Improving Lane Detection Using Integrated Camera 78

4.1. Lane Detection Method with Illumination Data 79

4.1.1. Image Data Quality Enhancements 79

4.1.2. Edge Extraction Improvements 80

4.1.3. Lane Tracking Improvements 81

4.1.3. Block Diagram for Proposed Lane Detection Method 83

4.2. Experimental Results 84

4.2.1. Design Results for Proposed Integrated Camera 84

4.2.2. Experiment Results with Real Road Conditions 88

4.2.3. Test Results for Angle of the Sun and Entering Tunnel 89

4.2.4. Test Result for Lane Detection 94

4.2.5. Test Results in Real Road Condition 99

5. Conclusions 103

Bibliography 106

List of Tables

TABLE I. Euro NCAP assessment protocol (scoring). 19

TABLE II. Measured results for light amount in plane surface. 30

TABLE III. Camera Specifications. 32

TABLE IV. Parameters Used in the Tests. 54

TABLE V. All Test Results for the Euro NCAP VRU Scenarios. 71

TABLE VI. Results of a Real Road Test (659 ㎞). 73

TABLE VII. CMOS sensor and dual light sensor. 87

TABLE VIII. Summary of detection of the Sun angle. 90

TABLE IX. Test results from real road. 102

List of Figures

Fig. 1.1. ADAS systems applied to Autonomous driving vehicle technology. 14

Fig. 1.2. Levels of autonomous driving technology. 16

Fig. 2.1. Edge detection. 24

Fig. 2.2. Filter for dark-bright-dark function. 24

Fig. 2.3. Filter for dark-bright-dark function. 25

Fig. 2.4. Block diagram for general lane detection algorithm. 27

Fig. 2.5. Circuit for dual light sensors. 28

Fig. 2.6. Illumination for dual light sensors. 29

Fig. 3.1. Configuration of the pedestrian AEB system. 34

Fig. 3.2. Control strategies for vehicle and pedestrian AEB. 35

Fig. 3.3. Pedestrian target selection algorithm. 38

Fig. 3.4. Concept of pedestrian target selection based on a funnel map. 40

Fig. 3.5. Impact of the prediction time. 41

Fig. 3.6. Predicted path of the vehicle. 43

Fig. 3.7. Lateral offset distance between a vehicle and the CIPP. 44

Fig. 3.8. Total distance to collision. 46

Fig. 3.9. Correlation of collision avoidance by braking or steering. 49

Fig. 3.10. Flowchart of the collision-warning judgement. 50

Fig. 3.11. Environment of the pedestrian AEB test. 52

Fig. 3.12. Front camera installed on the inner windshield. 52

Fig. 3.13. Test protocol of the Euro NCAP VRU. 56

Fig. 3.14. Test results for pedestrian target selection using analysis tool. 58

Fig. 3.15. Test results for CVFA (50%) scenario at 40 ㎞/h. 62

Fig. 3.16. Test results for CVNA-25 and CVNA-75 scenarios at 40 ㎞/h. 66

Fig. 3.17. Test results for the CVNC scenario at 40 ㎞/h. 70

Fig. 3.18. Typical images of urban situations. 75

Fig. 4.1. Improvement results of lane tracking. 82

Fig. 4.2. Block diagram for proposed lane detection algorithm. 83

Fig. 4.3. Actual design for circuit board of the proposed camera. 85

Fig. 4.4. Photograph of realized integrated front camera. 86

Fig. 4.5. The environment for the lane support system tests. 88

Fig. 4.6. The test results for detection of angle of the Sun. 90

Fig. 4.7. The test for tunnel entry detection. 92

Fig. 4.8. The test results for tunnel entry detection. 93

Fig. 4.9. Test results for image quality enhancements. 96

Fig. 4.10. Test results for edge extraction and tracking improvements. 98

Fig. 4.11. Real-road tests with various external situations. 99

Fig. 4.12. Real-road tests with different lane types. 101

초록보기

 Autonomous driving vehicle technology offers the innovation of changing transportation systems. Vehicles with this technology will likely maintain safety, reduce the energy consumption and pollution, and reduce the costs of congestion. Technological advancements are making a continuum between conventional, fully human-driven vehicles and autonomous driving vehicles, which partially or fully drive themselves and which may ultimately require no driver at all. This technologies has advanced to enable a vehicle to assist and make decisions for a human driver. Such technologies include adaptive cruise control (ACC) Stop & Go, lane keeping assist, lane change assist, autonomous emergency braking (AEB) and self-parking assist.

Recently, numerous vehicles have been installed with an AEB system for pedestrian protection. This system helps in avoiding or reducing accidents by alerting the driver and controlling the automatic brake actuator before an accident. Moreover, the Euro New Car Assessment Program (NCAP) has stipulated AEB pedestrian systems as a standard requirement since 2016. In addition, automotive companies studied the development of lane support systems in order to secure high scores on the NCAP. A front camera module and safety assistance systems are applied in intelligent vehicles. However, the front camera module has limitations in terms of backlight conditions, entering or exiting tunnels, and night driving because of lower image quality.

First, this dissertation presents pedestrian target selection using a funnel map for a pedestrian AEB system. The concept of target selection is based on crash probability calculations by comparing the pedestrian's predicted position and the current position to deduce the vehicle speed before an accident occurs. It is necessary to allow early breaking to avoid an accident. To determine the precise warning and brake timing, the distance to collision is calculated using the vehicle and sensor fusion information. The pedestrian target selection algorithm is tested using a real vehicle on a test track in three different scenarios for the Euro NCAP using a pedestrian dummy authorized by the Euro NCAP. Upon comparing the results before and after the application of the proposed algorithm, the longitudinal distance is shown to have a maximum margin of 1.7 m, and the vehicle speed has a maximum reduction effect of 11 km/h. Test results show that the proposed pedestrian AEB system can avoid or mitigate an accident when the vehicle travels at speeds up to 40 km/h.

Second, this dissertation proposes an integrated camera with a dual light sensor for improving the lane detection performance under the worst conditions. It includes a new algorithm to enhance the image data quality and improve the edge extraction and lane tracking using illumination information. The real-road tests are conducted under various external situation on an actual 728 km road and tests are collected by a total of 386,704 image frames. With the proposed idea, 373,542 image frames are correctly detected in tests. The test result shows that the false rate is reduced by 1.65 % using the proposed method as compared to the general method. The experiment results show that the system has a high potential in terms of reliability, enhancement, and improvements.

추천서가 (다양한 추천 자료를 만나보세요)

권호기사보기

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 기사목차

권호

기사명 원문보기 기사목차
닫기