Dynamic lane tracking system based on multi-model fuzzy controller

Large amounts of information can be obtained from road images by computer vision. This paper presents a vision-based lane detection algorithm to find the lane curves in each video frame, while a multi-model fuzzy controller is also established to fulfill lane following, which is based on fuzzy logic...

Full description

Saved in:
Bibliographic Details
Published in:International Conference on Industrial Mechatronics and Automation (Online) pp. 873 - 877
Main Authors: Qing Shi, Jin Zhao, Lei Han, Youjiang Ning, Guangwei Wang
Format: Conference Proceeding
Language:English
Published: IEEE 01.08.2016
Subjects:
Automation
Conferences
intelligent vehicle
lane detection
Mechatronics
multi-model control
ROI fitting
ISSN:2152-744X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Large amounts of information can be obtained from road images by computer vision. This paper presents a vision-based lane detection algorithm to find the lane curves in each video frame, while a multi-model fuzzy controller is also established to fulfill lane following, which is based on fuzzy logic. In the proposed detecting algorithm, a series of algorithms, including image preprocessing and lane extraction algorithm, are done to extract the edge features, then lane-fitting is done successfully inside the region of interest(ROI) based on the coordinate transformation. Furthermore, some real-time simulation tests had been done successfully for verification, and the lateral position error was calculated. Meanwhile, the multi-model fuzzy controller, which inherits the advantages of both the multi-model control and fuzzy control, has been used at a low speed. In the end, the integrated simulation results validate the good tracking performance of this algorithm.
ISSN:2152-744X
DOI:10.1109/ICMA.2016.7558677