Experimental Studies on Dynamics Performance of Lateral and Longitudinal
Control for Autonomous Vehicle Using Image Processing
Khalid Isa
Department of Computer Engineering
Faculty of Electrical and Electronic Engineering
Universiti Tun Hussein Onn Malaysia
P.O Box 101, 86400 Parit Raja, Batu Pahat, Johor, Malaysia
halid@uthm.edu.my
Abstract
This paper presents a simulation of vehicle driving
control system in terms of lateral and longitudinal
control using image processing. The main contribution
of this study is it contributes an algorithm of vehicle
lane detection and tracking which based on colour cue
segmentation, Canny edge detection and Hough
transform. The algorithm gave good result in detecting
straight and smooth curvature lane on highway even
the lane was affected by shadow. Then by combining
and processing the result of lane detection process
with vehicle dynamics model, this system will produce
the dynamics performance of vehicle driving control.
This simulation system was divided into four
subsystems: sensor, image processing, controller and
vehicle. All the methods have been tested on video data
and the experimental results have demonstrated a fast
and robust system.
1. Introduction
The main objective of this system is to develop the
simulation for analysing dynamics performance of
lateral and longitudinal control for autonomous vehicle
using image processing. Therefore, the simulation
determines the steering command for the vehicle lateral
control by processing, analysing, and detecting the lane
on highway. This means, the lane detection process
will produce the lane angle, and this angle was directly
used as steering command. Then, by combining the
steering command and others vehicle dynamics
parameters such as the vehicle mass, and vehicle
velocity, the vehicle’s dynamics performance can be
determined by this system.
Previously, many lanes or road boundary detection
algorithms have been developed. LOIS [1] system used
a deformable template approach to find the best fit of
road model whether it straight or curve. The research
groups of the University Der Bundeswehr [2] and
Daimler-Benz [3] base their road detection
functionality on a specific road model: lane markings
are modelled as clothoids. This model has the
advantage that the knowledge of only two parameters
allows the full localization of lane markings and the
computation of other parameters like the lateral offset
within the lane, the lateral speed with respect to the
lane and the steering angle. The approach in [4] is an
evolutionary approach of lane markings detection. It
used collaborative autonomous agents to identify the
lane markings in road images.
Several aspects of designing control system for a
vehicle have been examined extensively in the past,
both in the physics literature [5] as well as in control
theoretic studies. The control problem in a dynamic
setting, using measurement ahead of the vehicle, has
been explored by [6], who proposed a constant control
law proportional to the offset from the centreline at a
look-ahead distance. Ackermann et al [7], proposed a
linear and non-linear controller design for robust
steering. Taylor et al [8] considered the problem of
controlling a motor vehicle based on the information
obtained from conventional cameras mounted onboard.
Ma, Kosecka and Sastry [9] looked at the problem of
guiding a nonholonomic robot along a path based on
visual input.
2. Problem Formulation
The following subsection presents the basic system
design and the techniques.
2.1 System Design
System design for autonomous vehicle is depending
on number of tasks that can be performed by the
vehicle. Since the experiment only using one video
camera as a sensor, so this paper only presented the
lane detection task along with dynamics and control of
IEEE 8th International Conference on Computer and Information Technology Workshops
978-0-7695-3242-4/08 $25.00 © 2008 IEEE
DOI 10.1109/CIT.2008.Workshops.89
411