Article
Lane Detection Algorithm Using LRF for Autonomous
Navigation of Mobile Robot
Jong-Ho Han
1
and Hyun-Woo Kim
2,
*
Citation: Han, J.-H.; Kim, H.-W.
Lane Detection Algorithm Using LRF
for Autonomous Navigation of
Mobile Robot. Appl. Sci. 2021, 11,
6229. https://doi.org/
10.3390/app11136229
Academic Editor: Dario Richiedei
Received: 20 April 2021
Accepted: 2 July 2021
Published: 5 July 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
1
Test & Evaluation Division, Korea Intelligent Automotive Parts Promotion Institute, 201, Gukgasandanseo-ro,
Guji-myeon, Daseong-gun, Daegu 43011, Korea; jonghohan@kiapi.or.kr
2
Marine IT Convergence Materials Research Division, Research Institute of Medium & Small Shipbuilding,
38-6, Noksan industrial Complex 232-ro, Gangseo-gu, Busan 46757, Korea
* Correspondence: kimhw@rims.re.kr; Tel.: +82-051-974-5549
Abstract:
This paper proposes a lane detection algorithm using a laser range finder (LRF) for the
autonomous navigation of a mobile robot. There are many technologies for ensuring the safety of
vehicles, such as airbags, ABS, and EPS. Further, lane detection is a fundamental requirement for an
automobile system that utilizes the external environment information of automobiles. Representative
methods of lane recognition are vision-based and LRF-based systems. In the case of a vision-based
system, the recognition of the environment of a three-dimensional space becomes excellent only
in good conditions for capturing images. However, there are so many unexpected barriers, such
as bad illumination, occlusions, vibrations, and thick fog, that the vision-based method cannot be
used for satisfying the abovementioned fundamental requirement. In this paper, a three-dimensional
lane detection algorithm using LRF that is very robust against illumination is proposed. For the
three-dimensional lane detection, the laser reflection difference between the asphalt and the lane
according to color and distance has been utilized with the extraction of feature points. Further, a
stable tracking algorithm is introduced empirically in this research. The performance of the proposed
algorithm of lane detection and tracking has been experimentally verified.
Keywords: lane detect; tracking; 3D map; real time; laser range finder; curvature
1. Introduction
There has been considerable progress in the field of vehicle safety systems in the last
few decades—from safety belts in the 60s to electrical systems such as air bags, anti-lock
brake systems (ABSs), and electric power steering (EPS) in the 90s; this progress has been
aimed at increasing passenger safety. However, these systems offer only passive safety
because they operate in response to the state of the in-vehicle system [
1
]. In line with
the recent development of smart technologies, considerable research has been conducted
on the implementation of intelligent and sophisticated safety systems [
2
]. Such a system
is referred to as an advanced safety vehicle (ASV). An ASV is a safety system designed
to alarm drivers in advance against accident-prone situations so that they can cope with
such situations actively. ASV-enabled vehicles are cutting-edge, safe vehicles equipped
with artificial intelligence involving various types of intelligent safety technologies and
thus maximizing driving safety and comfort. ASV’s representative core technologies are
a vehicle collision alarm system and a lane recognition system. The former is a vehicle-
to-vehicle distance detection sensor that alarms the driver when the distance from the
vehicle in front or behind becomes alarmingly small. The latter, commonly called a lane
departure warning system (LDWS), is a sensor detecting the vehicle leaving its lane; it
checks the lane and informs the driver of an imminent lane departure with an audible
signal. The sensors mainly used in ASV are an ultrasonic sensor to recognize obstacles
in front, a vision-based system to recognize the lane and road situations such as traffic
lights with radar, and a laser range finder (LRF). The information detected is used for
Appl. Sci. 2021, 11, 6229. https://doi.org/10.3390/app11136229 https://www.mdpi.com/journal/applsci
