Autonomous Lane Keeping System: Lane Detection, Tracking and Control on Embedded System
- PDF / 1,754,404 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 17 Downloads / 264 Views
ORIGINAL ARTICLE
Autonomous Lane Keeping System: Lane Detection, Tracking and Control on Embedded System Mingjie Liu1 · Xutao Deng1 · Zhen Lei1 · Chao Jiang2 · Changhao Piao1 Received: 6 November 2019 / Revised: 8 May 2020 / Accepted: 1 October 2020 © The Korean Institute of Electrical Engineers 2020
Abstract Autonomous lane keeping system is the key technique to autonomous driving. It includes lane detection, lane tracking and control. It has been developed enormously, but it is still a challenge work due to different factors such as illumination, general hyper-parameters setting for different road condition and lane boundary correction. In addition, due to imbalance on accuracy and processing time, it is hard to conduct in embedding system. In this study, an autonomous lane keeping system is developed based on deep learning. First, a lane detection and tracking system is designed, which is robust to lane boundary correction. Especially for lane detection, a light-weight network named as LaneFCNet is proposed, which can balance accuracy and processing time. Then, lane tracking was followed by detector to improve the detection performance and create autonomous driving trajectory. Finally, to brief lane fitting problem, it was treated as ridge regression problem, which can enhance the effectiveness to the whole system. Experimental results show that our integrated lane detection and tracking system can trade off accuracy and processing time and the whole line keeping system is robust enough to autonomous driving. Keywords Autonomous driving · Autonomous lane keeping system · Lane detection and tracking system
1 Introduction Autonomous lane keeping system, which includes lane detection, lane tracking and control, is the key techniques in autonomous vehicles and driver support systems, especially in advanced driving assistant system (ADAS) [1–3]. Lane detection and tracking, as fundamental problems in computer vision tasks, highly affect the performance of * Chao Jiang [email protected] Mingjie Liu [email protected] Xutao Deng [email protected] Zhen Lei [email protected] Changhao Piao [email protected] 1
College of Automation, Chongqing University of Posts and Telecommunications, No. 2 Chongwen Road, Chongqing, China
Department of Mechanical Engineering, Inha University, 100 Inha‑ro Nam‑gu, Incheon, Republic of Korea
2
autonomous lane keeping system. Lane detection is the problem of locating lane boundaries without knowledge of the road geometry. Normally, it is followed by lane tracking, which can enhance the detection performance. In the vision-based lane detection methods, it can be divided into two categories: feature extraction-based method and modelbased method. It describes the lane structure by mathematic model with some hyper-parameters in model-based method. It can resist the effect of the noise; however, it is difficult to set the adaptive hyper-parameters and hard to realize. The lane structure is described by features such as Hough transform in traditional feature extraction-
Data Loading...
