Development of Steering Control System for

Autonomous Vehicle Using Geometry-Based

Path Tracking Algorithm

Myungwook Park, Sangwoo Lee, and Wooyong Han

In this paper, a steering control system for the path I. Introduction

tracking of autonomous vehicles is described. The steering
control system consists of a path tracker and primitive With advancements in information and communication
driver. The path tracker generates the desired steering and vehicle techniques, the competition to develop and
angle by using the look-ahead distance, vehicle heading, commercialize intelligent and autonomous vehicles has
and a lateral offset. A method for applying an autonomous intensified. The aim of these vehicles is to secure vehicle
vehicle to path tracking is an advanced pure pursuit stability and driver ease, as well as to enhance traffic efficiency.
method that can reduce cutting corners, which is a The intelligent Advanced Driver Assistance System (ADAS)
weakness of the pure pursuit method. The steering helps drivers through a human–machine interface by following
controller controls the steering actuator to follow the the lane departure warning system, Lane Keeping Assistance
desired steering angle. A servo motor is installed to control System (LKAS), Forward Collision Warning System, and
the steering handle, and it can transmit the steering force Smart Parking Assistant System (SPAS). LKAS helps a car
using a belt and pulley. We designed a steering controller stay on course near the center of a lane by continuously
that is applied to a proportional integral differential applying a small amount of counter-steering force when the
controller. However, because of a dead band, the path system senses the vehicle deviating from its lane [1]. SPAS
tracking performance and stability of autonomous helps make parking easier by providing automatic steering
vehicles are reduced. To overcome the dead band, a dead assistance along a preset path to guide a driver to the optimal
band compensator was developed. As a result of the position from which to start backing into a parking space [2].
compensator, the path tracking performance and stability An autonomous vehicle is a vehicle that can drive itself to a
are improved. destination without its driver having to recognize the driving
environment. This paper describes the method of path tracking
Keywords: Autonomous vehicle, steering control system, and the development of a steering control system using
geometry path-tracking method. techniques applied to intelligent and autonomous vehicles.
The geometric path-racking method is one of the most
Manuscript received Aug. 10, 2014; revised Mar. 6, 2015; accepted Apr. 21, 2015. popular methods, following pure pursuit, vector pursuit, and
This work was supported by Institute for Information & Communications Technology the Stanley method. The pure pursuit method is relatively easy
Promotion (IITP) grant funded by the Korea government (MSIP) (No. B0101-15-0134,
Development of Decision Making/Control Technology of Vehicle/Driver Cooperative
to implement and is robust to large disturbances. However, this
Autonomous Driving System Based on ICT). method has a problem of cutting corners at high speed, because
Myungwook Park (corresponding author, [email protected]), Sangwoo Lee (swlee@ the look-ahead distance increases along with an increase in
etri.re.kr), and Wooyong Han ([email protected]) are with the IT Convergence Technology
Research Laboratory, ETRI, Daejeon, Rep. of Korea. vehicle speed. In addition, steady-state errors occur at high

ETRI Journal, Volume 37, Number 3, June 2015 © 2015 Myungwook Park et al. 617
http://dx.doi.org/10.4218/etrij.15.0114.0123
speed. The Stanley method has a better tracking performance Velodyne
GPS antenna
than the pure pursuit method under most driving environments.
In addition, because it uses the lateral offset from the front axle Steering Camera INS solution
actuator
of the vehicle to the path, the Stanley method does not incur the Vehicle
controller
problem of cutting corners but does incur an overshooting of
turns. It is also influenced by various disturbances [3]. The
77 GHz
vector pursuit method uses the theory of screws. The vector radar
pursuit method is similar to a pure pursuit method. However,
this method is less sensitive regarding the look-ahead distance IMU
than the pure pursuit method [4]. The pure pursuit method is Multi-layer
laser scanner
used to apply path tracking to an autonomous vehicle, is easy
to implement, and is robust to large disturbances. In addition, Brake APS signal Gear shift DMI 2D laser
actuator actuator scanner
we propose an algorithm to reduce the problem of cutting
corners that occurs in the pure pursuit method by using a lateral Fig. 1. Hardware system of autonomous vehicle.
offset from the rear axle of the vehicle to the path.
To follow a desired steering angle generated by the path dc convertors. In addition, the alternator is installed in the
tracking algorithm, we have to control the steering handle. vehicle engine room for charging the battery. The perception
Today, the steering system found in most vehicles is that of module recognizes the driving environment and detects objects
the motor-driven power steering (MDPS) system. The MDPS and drivable paths using a 2D/3D laser scanner, 77 GHz radar,
system reduces a driver’s effort by providing assist torque and and a camera. The decision-making module generates the
return torque based on an electric motor. Recently, the MDPS reference path of the vehicle based on the recognition results.
system has been applied to the ADAS and autonomous The control module controls the vehicle to track the reference
systems, and beyond its original functions, stands in for a driver path. For path tracking, the steering actuator, brake actuator,
[5]. However, for security reasons, we cannot control the and gear shift actuator are equipped in the vehicle. In addition,
MDPS system. Therefore, to control the steering handle, we to know the moving distance and vehicle speed, two distance
installed an actuator on the steering column. In addition, a measurement instruments (DMIs) are implemented in the rear
proportional integral differential (PID) controller and dead wheels. Because our system is an autonomous system based on
band compensator are applied to control the actuator, and we the global position of the vehicle, an inertia navigation system
verified the controller through a field test. (INS) based on a virtual reference station is installed
The remainder of this paper is outlined as follows. Section II to estimate the vehicle position and dynamics [6]–[7]. The
describes our hardware systems. Section III introduces an hardware system of the autonomous vehicle is shown in Fig. 1.
overview of the steering control system. The advanced pure
pursuit algorithm is then described in Section IV. Section V 2. Steering System
then presents a primitive driver to control the steering system.
The experimental results are provided in Section VI. Finally, The steering system of most vehicles is an MDPS system.
some concluding remarks and areas of future work are given in An MDPS system reduces a driver’s effort by providing assist
Section VII. torque and return torque using an electric motor. In addition, it
is possible to maintain the lane, park automatically, and drive
autonomously using the MDPS system. However, for security
II. Hardware Systems reasons, we cannot control the MDPS system. Therefore, to
track the reference path of the autonomous vehicle, it is
1. Autonomous Vehicle
possible to control the steering handle by mounting an actuator
We use a commercial sport utility vehicle for applying the and reduction gear on the steering column, where the actuator
lateral control system. Because this vehicle has a lot of space, it transmits the torque using a belt and pulley. In addition, a
is easy to install the power system, controllers, sensors, and potentiometer is equipped to measure the motor position on the
actuators used in the autonomous system. motor shaft. These actuators and sensors communicate with the
An autonomous vehicle consists of a power module, controller using an RS232 and controller area network (CAN).
perception module, decision-making module, and control The controller reads the position and speed of the motor from
module. We used a 24 V/100 Ah battery for the power module, the CAN bus, and it then writes the desired steering PWM
which supplies power to each of the other modules using dc-to- signal from the RS232. The hardware system configuration of

618 Myungwook Park et al. ETRI Journal, Volume 37, Number 3, June 2015
http://dx.doi.org/10.4218/etrij.15.0114.0123
 Position Steering control system
sensor Steering Path tracker Primitive driver
CAN column
Path V_des
12 V Velocity planning Throttle/brake
Belt & pulley controller
BMS Decision look-ahead distance Vehicle
24 V INS data
δ_des Motor
RS232 Path tracking controller
Controller

Motor &
reduction gear Feedback data
MDPS
Fig. 4. Block diagram of steering control system.

steering controller. The steering controller controls the steering

actuator to follow the reference path. In this paper, we do not
Fig. 2. Hardware configuration of steering control system. deal with a throttle/brake controller. The steering control
system is shown by the dotted line in Fig. 4.

IV. Path Tracker

1. Velocity Planning

Velocity planning is needed to prevent slipping and roll-

overs when the autonomous vehicle system is running. For this,
the velocity planning module plans the velocity using the
curvature of the path, the side friction factor, and the super-
elevation.
(a) (b)
A vehicle driving on a horizontally curved road needs
Fig. 3. (a) Servo motor and position sensor installed in autonomous
centripetal force for circular motion. Depending on the circular
vehicle’s steering column and (b) steering controller.
motion, the horizontal curve radius, and speed, the centrifugal
force (see Fig. 5) is applied through (1) below. In (1), g is the
the steering control system is shown in Fig. 2, whereas the
gravitational acceleration, v is the velocity, R is the radius of the
motor sensor installed in the steering column and the controller
curvature, and w is the weight of the vehicle.
are shown in Fig. 3.
w v2
f  . (1)
III. Steering Control System g R

If the side friction factor is f for the lateral friction force, for
The steering control system is a path tracking system that
vehicle safety, then (2) below is satisfied.
controls the steering actuator based on the current vehicle
position, heading from the INS, and reference path from the (f cos   w sin  )  f (f sin   w cos  ), (2)
path planner. The steering control system consists of a path
where θ is the bank angle. Dividing both sides of (2) by cos θ,
tracker and primitive driver. The path tracker consists of the
we obtain
following three modules: a velocity planning module, look-
ahead distance decision module, and path tracking module. (f  w tan  )  f (f tan   w ). (3)
The velocity planning module plans the velocity using
the curvature of the path, the side friction factor, and the Using super-elevation i = tan θ, substituting (1) into (3), and
super-elevation. The look-ahead distance decision module dividing both sides by w, we obtain the following equation:
determines the look-ahead distance depending on the velocity  v2   v2 
of the vehicle. In addition, after selecting the goal point on the  i  f  i  1 . (4)
 g  R   g  R 
reference path based on the look-ahead distance, the path
tracking module generates the desired steering angle. The Assuming i·f is zero in (4), the radius of horizontal curvature is
primitive driver consists of a throttle/brake controller and given by

ETRI Journal, Volume 37, Number 3, June 2015 Myungwook Park et al. 619
http://dx.doi.org/10.4218/etrij.15.0114.0123
 Streets [8].
The side friction factor is defined based on the vehicle speed,
quality of the road material, and the road conditions (dry, wet,
f or icy road surface), and is limited to 0.1 to 0.16 depending on
the vehicle velocity. The result of the speed according to the
w curvature is shown in Fig. 6, where the maximum velocity is
θ R limited to 60 kph.

Fig. 5. Acting centrifugal force. 2. Determination of Look-Ahead Distance

The magnitude of the look-ahead distance is dynamically
60
adjusted based on the vehicle velocity. A small magnitude of
55
50 look-ahead distance forces the system to track the path more
45 accurately, and increases in the maximum curvature of a path
Vel_des (kph) & Curve×100

40
35 can be tracked. Additionally, a small look-ahead distance
30 results in oscillation while the vehicle is following the path.
25
20 Because of this, the vehicle can become unstable. On the other
15 hand, a large magnitude reduces oscillations while tracking the
10 Vel_des
5 Curve path and allows the vehicle to begin turning before it reaches
0 the curve, resulting in smoother trajectories around sharply
–5
–10
curved paths.
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000
In addition, a large magnitude reduces overshoot as the look-
Fig. 6. Desired velocity using curve. ahead distance acts as a damping factor. But if it is too large,
then the tracking performance will deteriorate, such as in the
cutting corners phenomenon [9].
v2 1 v cur  10 kph,
R  . (5) 5
g i f 
ld  0.5  v cur 10 kph  v cur  50 kph, (8)
Substituting 1/κ for R, the limit velocity at a horizontally  50 kph  v cur .
25
curved road is calculated through the following equation:
S.F. Campbell defined the look-ahead distance as a function of
g  (i  f ) the vehicle velocity and verified that the driving stability of the
v , (6)
 vehicle increases if the look-ahead distance increases [10]. In this
paper, the look-ahead distance is set to between 5 m and 25 m
where ĸ is the curvature of the road.
within a velocity range of 10 kph to 50 kph, as shown in Fig. 7.
To find the curvature, it is assumed that the path is curved on
a plane. In addition, the path is assumed to be approximable by
3. Advanced Path Tracking Algorithm
a third-degree polynomial. After approximating a desired
path with a third-degree polynomial with curve fitting, the The pure pursuit algorithm consists of geometrically
curvatures of the polynomial were determined. From the
definition of the curvature, the curvature of the polynomial y = ld (m)
x(t) is shown below.
d2 y 25

 dx 2 , (7) 20
3/ 2
  dy  
2

1     15
  dx   10
The road super-elevation is designed to be 6%–8% when a
5
road is built. In addition, the road super-elevation is limited to
vcur (kph)
below 6% in local and urban roads according to the American 0 5 10 15 20 25 30 35 50
Society of State Highway and Transportation Officials
publication, A Policy on Geometric Design of Highways and Fig. 7. Look-ahead distance as function of speed.

620 Myungwook Park et al. ETRI Journal, Volume 37, Number 3, June 2015
http://dx.doi.org/10.4218/etrij.15.0114.0123
calculating the curvature of a circular arc that connects the
center of the rear axle to the goal point on the path ahead of the
vehicle. The goal point is determined from the look-ahead
distance (ld). The desired steering wheel angle of the vehicle is
2
determined using angle α between the heading angle of the R

vehicle and the look-ahead vector. The pure pursuit geometry is

shown in Fig. 8.
R
ld R
 , (9)
sin(2 )  π 
sin     Reference (gx, gy)
2  path
ey
ld R  δc
 , (10)
2sin( ) cos( ) cos( ) Circular arc

ld ld
 2 R, (11)
sin( )

2sin( ) Fig. 8. Pure pursuit geometry.

 , (12)
ld
where ĸ is the curvature of the circular arc. Applying an
Ackerman model with two degrees of freedom, the steering
angle is given as
  tan 1 ( L), (13) Goal point
Vehicle trajectory
where L is the wheel base. Using (12) and (13), the desired Reference path
steering angle can be written as
 2 L sin  (t )  
 l (t )  tan 1  . (14)
d
 ld  ey ey
The pure pursuit is influenced by the look-ahead distance,
which is dynamically adjusted with the velocity. A
characteristic of the pure pursuit is that a sufficient look-ahead
Fig. 9. Examples of lateral offset (left: cutting corners and right:
distance will result in cutting corners while tracking a curved steady-state error).
path. To improve this characteristic and steady-state error, we
use the lateral offset (error) between the vehicle and path. The
examples of lateral offset are shown in Fig. 9. To the left is the  e (t )  ey (t )   ( (t )) ey dt.
y
(15)
cutting corners phenomenon, and to the right is the steady-state
The integral gain is a function of the road curvature.
error. To reduce the lateral offset, we apply proportional-
The total desired steering angle, δt, is given through the
integral (PI) control theory, as in equation (15) below.
following equation:
The desired steering angle according to the lateral offset
 ey (t ) is the sum of the proportional term ey (t ) and  t l  e ,
d y
(16)
integral term  ( (t )) ey dt , where  and Q(κ(t)) are the where  ld is the desired steering angle by look-ahead distance.
proportional and integral gains, respectively; both of which are
positive values. V. Primitive Driver
The proportional term performs the role of reducing the
lateral offset by reducing the target steering in the cutting The primitive driver as a low-level controller drives the
corners phenomenon. On the other hand, the integral term is actuators to move the autonomous vehicle. The primitive
useful when a path has a small curvature and eliminates the driver consists of a throttle/brake controller and steering
residual steady-state error. controller.

ETRI Journal, Volume 37, Number 3, June 2015 Myungwook Park et al. 621
http://dx.doi.org/10.4218/etrij.15.0114.0123
1. Steering Controller
δdes (deg) δcur (deg) Tmotor (%) yerror (m) Dead band
10
The steering controller controls the steering actuator to
follow the desired steering angle.
0
A. PID Controller 6% 2s

The advantages of a PID controller are its easy design and –10
good performance. Thus, a PID controller has been used in 86 90 95 100 105
Time (s)
many applications. Figure 10 shows a block diagram of the
steering controller using a PID controller. The error, the Fig. 11. Experimental results of steering controller using PID
difference between the desired steering angle and actual controller.
steering angle, is sent to the PID controller, and the controller
computes the pulse-width modulation (PWM) signal. This
Motor torque (Nm)
signal is sent to the motor, and the steering motor controls the
steering handle using the output torque. A mathematical
description of a typical PID controller is shown in (17).
Dead band
t
de
MV(t )  K p e(t )  K i e( )d  K d , (17) Error (m)
0 dt
where Kp, Ki, and Kd are the proportional gain, integral gain,
and derivative gain, respectively, which we determine
experimentally [11]. Fig. 12. Dead band compensator.
We verified the steering controller designed using a PID
controller on a straight asphalt road. In this test, a longitudinal δdes (deg) δcur (deg) Tmotor (%) yerror (m) Dead band
speed of 40 kph is used. Figure 11 shows the results. The red 7

line in the figure is the desired steering angle, the black dotted 0.2 s
line is the current steering angle, the green dotted line is the 6%
0
percent of torque value, and the blue line is the lateral offset.
Checking the actual steering angle data, we can see sections
without a variation for a short period of time in spite of the –7
motor torque output. In such sections, because the motor torque 26 27 28 29 30 31
Time (s)
value is small, the movement of the steering handle is not
influenced by the steering motor. This is called the dead band, Fig. 13. Experimental results of steering controller using PID
controller and dead band compensator.
and it occurs within a range of –6% to 6% torque. As a result, it
generates a time delay of about 2 s when following the desired
steering angle. In spite of the straight road and low speed, a
lateral offset occurs at more than 0.3 m, and the more the δ_des + Error
PID
controller PWM Torque
Turning
angle
Steering
Servo motor Vehicle
vehicle velocity increases, the more the lateral error increases. – Dead band
Σ
handle

The reason for the dead band is the hardware characteristic compensator
Current steering angle
of the steering system. Because the developed steering system
transmits the torque using a toothed belt and pulley, some Fig. 14. Block diagram of steering controller using PID controller
and dead band compensator.
physical elements exist, such as a backlash, which interfere
with the controls. To fix this problem, we designed a dead band
compensator.
Turning
δ_des + Error PID PWM Servo Torque Steering angle B. PID Controller and Dead Band Compensator
Vehicle
– controller motor handle
To reduce the time delay and lateral error, in this paper,
Current steering angle
we designed a dead band compensator (see Fig. 12). To
immediately follow the desired angle, the dead band
Fig. 10. Block diagram of steering controller using PID controller.
compensator compensates the minimum torque needed to

622 Myungwook Park et al. ETRI Journal, Volume 37, Number 3, June 2015
http://dx.doi.org/10.4218/etrij.15.0114.0123
move the steering handle. The sign of the compensated torque
depends on the sign of the error. If the error of the block
diagram is positive, then the compensated torque is about 6%.
If the error of the block diagram is negative, the compensated
torque is about –6%, as shown in Fig. 13. The final desired Fig. 15. Test road: typical asphalt road.
output value is the sum of the output of each controller.
Figure 14 shows a block diagram of the steering controller δdes (deg) δcur (deg) Tmotor (%) yerror (m)
6

Steering angle & motor torque

using a PID controller and dead band compensator.
We verified the steering controller designed using a PID
controller and dead band compensator. In our test, a 0
longitudinal speed of 80 kph is used. Figure 13 shows the
results. As shown in the graph, the dead band section still exists,
–6
but the time delay decreases from 2 s to 0.2 s. In addition, the 0.4

improving controller reduces the lateral error by about 0.24 m.

Lateral error
0 0.24 m

VI. Experimental Results –0.4

20 30 40 50
Time (s)
To verify the proposed steering control system, autonomous
vehicles were tested on a typical asphalt road, as shown in Fig. 16. Experimental results of lane following maneuver at test
velocity of 80 kph.
Fig. 15. The distance of the test road is about 800 m, and the
road has two lanes, where the width of each lane is 3.5 m in
each direction. We conducted two different types of δdes (deg) δcur (deg) Tmotor (%) yerror (m)
Steering angle & motor torque

6
experiments to verify the steering control system. One is a lane
following maneuver, and the other is a double change
maneuver. In these tests, longitudinal speeds of 80 kph and 0

100 kph were used.

–6
1. Lane Following Maneuver 0.4
Lateral error

–0.24 m–0.3 m
The experimental environment consists of a straight section 0

and curved section with a road curvature of 0.15. Figures 16

–0.4
and 17 show the experimental results of the lane following 18 20 25 30 35
Time (s)
maneuver.
Because of the influence of the dead band, the performance Fig. 17. Experimental results of lane following maneuver at test
velocity of 100 kph.
of the steering controller is not perfect, but the path tracking
performance is good. The maximum lateral errors are about
0.24 m and 0.3 m in each longitudinal velocity in spite of the
high velocity. 3.5 m

2. Double Lane Change Maneuver A B A

The experimental environment for this maneuver consists of Fig. 18. Schematization of test road for double lane change
a straight section and curved section with a road curvature of maneuver.
0.15.
The path of a double lane change maneuver is schematized 166 m at 100 kph. Thus, during this maneuver, the lengths of
in Fig. 18. The experimental environment in this maneuver lane change section A were set at 100 m and 150 m. In addition,
consists of a straight section and a lane change section. For the lengths of section B were set at 100 m and 150 m.
highways, the mean lane change durations range from 5.14 s to Figures 19 and 20 show the experimental results of the double
6 s [12]–[13]. The lane change lengths are calculated to be lane change maneuver using a path length of section A of
between 114 m and 133 m at 80 kph, and between 138 m and 100 m. The performance of the steering controller is similar to

ETRI Journal, Volume 37, Number 3, June 2015 Myungwook Park et al. 623
http://dx.doi.org/10.4218/etrij.15.0114.0123
 δdes (deg) δcur (deg) Tmotor (%) yerror (m)
9 δdes (deg) δcur (deg) Tmotor (%) yerror (m)
Steering angle & motor torque

Steering angle & motor torque

6

0
0

A B A
–9
0.6 –6 A B A
0.4
Lateral error

Lateral error
0 –0.56 m –0.58 m
0 –0.48 m–0.3 m
–0.6
11 15 20 25 30 35 –0.4
Time (s) 18 20 25 30 35 37
Time (s)
Fig. 19. Experimental results of double lane change for vehicle
velocity of 80 kph and lane change section length of Fig. 22. Experimental results of double lane change for vehicle
100 m. velocity of 100 kph with lane change section length of
150 m.

δdes (deg) δcur (deg) Tmotor (%) yerror (m)

9 maximum lateral error is about 0.58 m at 80 kph. In Fig. 20, the
Steering angle & motor torque

maximum lateral error is about 1 m at 100 kph. As a result of a

large lateral error, the vehicle departed the lane.
0
Figures 21 and 22 show the experimental results of a double
A B A lane change maneuver using a path length of of 150 m. The
–9 performance of the steering controller is similar to the results of
1
other experiments. In addition, overshoots occurred. However,
Lateral error

0 0.95 m
because the length of the lane change increased, the overshoot
was reduced by 50%. The results show that the maximum
–1 lateral errors are about 0.32 m and 0.48 m at each velocity. As a
15 20 25 30 35 37
Time (s) result of the double lane change maneuver, if the lane change
Fig. 20. Experimental results of double lane change for vehicle length increases, then the vehicle can stably change lanes.
velocity of 100 kph and lane change section length of
100 m.
VII. Conclusion

δdes (deg) δcur (deg) Tmotor (%) yerror (m)

For this paper, we developed a steering control system for
the path tracking of an autonomous vehicle. A servo motor was
Steering angle & motor torque

installed to control a steering handle, and it can transmit the

steering force using a belt and pulley. The steering control
0
system is a path tracking system that controls the steering
actuator based on the current vehicle position, heading from the
–6 A B A INS, and reference path from the path planner. The method
0.4
for applying an autonomous vehicle to path tracking is an
Lateral error

0
–0.25 m–0.32 m advanced pure pursuit method that can reduce cutting corners,
which is a weakness of the pure pursuit method.
–0.4 We designed the steering controller to apply to a PID
18 20 25 30 35 40 45
Time (s) controller using an error; that is, the difference between the
Fig. 21. Experimental results of double lane change for vehicle desired steering angle and actual steering angle. However,
velocity of 80 kph and lane change section length of because of some physical elements, the dead band is a finite
150 m. range in which the system cannot detect the output variation.
Based on the influence of the dead band, the performance of
the results of the lane following maneuver. However, the the path tracking and stability of the autonomous vehicle are
overshoots occur immediately after the autonomous vehicle reduced. We developed a dead band compensator to overcome
enters and exits the lane change section. In Fig. 19, the a dead band, and improved the path tracking performance and

624 Myungwook Park et al. ETRI Journal, Volume 37, Number 3, June 2015
http://dx.doi.org/10.4218/etrij.15.0114.0123
stability. Technology, Cambridge, USA, 2007.
Two different kinds of experiments were conducted to verify [11] K.H. Ang, G. Chong, and Y. Li, “PID Control System Analysis,
the steering control system. In the lane following maneuver, the Design, and Technology,” IEEE Trans. Contr. Syst. Technol., vol.
lateral error was 0.3 m or less in spite of a high velocity. In a 13, no. 4, July 2005, pp. 559–576.
double lane change maneuver, an overshoot occurred for a [12] J.E. Naranjo et al., “Lane-Change Fuzzy Control in Autonomous
short lane change length. When the lane change length Vehicles for the Overtaking Maneuver,” IEEE Trans. Intell.
increases, stable driving of the vehicle is possible. Transp. Syst., vol. 9, no. 3, Sept. 2008, pp. 438–450.
In this paper, we assumed that there is no influence on the [13] L.-S. Jin et al., “Research on Safety Lane Change Model of
disturbances (crosswind, driver torque) and did not apply tire or Driver Assistant System on Highway,” IEEE Intell. Vehicles
road models. For this reason, the lateral errors do not converge Symp., Xi’an, China, June 3–5, 2009, pp. 1051–1056.
near to zero.
In our future work, to reduce the tracking error, we need to Myungwook Park received his BS degree in
develop a model-based algorithm that is independent of the mechanical engineering from the Engineering
curvature of the road. In addition, more experiments are College, Kookmin University, Seoul, Rep. of
required to verify the system under various environments. Korea, in 2006 and his MS and PhD degrees in
electronic control systems from the School of
References Automotive Engineering, Kookmin University,
in 2008 and 2012, respectively. Since 2013, he
[1] O. Tunçer et al., “Vision Based Lane Keeping Assistance Control has been with the Electronics and Telecommunications Research
Triggered by a Driver Inattention Monitor,” IEEE Int. Conf. Syst., Institute, Daejeon, Rep. of Korea, where he is now a senior researcher.
Man, Cybern., Istanbul, Turkey, Oct. 10–13, 2010, pp. 289–297. His main research interests are autonomous vehicles, vehicle electronic
[2] Y. Wang and X. Zhu, “Design and Implementation of an control systems, and control theory.
Integrated Multi-functional Autonomous Parking System with
Fuzzy Logic Controller,” American Cont. Conf., Montreal, Sangwoo Lee received his BS and MS degrees
Canada, June 27–29, 2012, pp. 2928–2933. in electronic communication engineering from
[3] J.M. Snider, “Automatic Steering Methods for Autonomous the Engineering College, Kwangwoon
Automobile Path Tracking,” M.S. thesis, Robot. Institute, University, Seoul, Rep. of Korea, in 1994 and
Carnegie Mellon Univ., Pittsburgh, PA, USA, Feb. 2009. 1996, respectively. From 1996 to 2000, he
[4] J.S. Wit, “Vector Pursuit Path Tracking for Autonomous Ground worked for Daewoo Electronics, Seoul, Rep. of
Vehicles,” Ph.D. dissertation, Dept. Mechanical and Aerospace Korea. Since 2000, he has been with the
Eng., Univ. of Florida, Gainesville, USA, 2000. Electronics and Telecommunications Research Institute, Daejeon, Rep.
[5] A. Sugiyama et al., “An EPS Control Strategy to Reduce Steering of Korea, where he is now a principal researcher. His main research
Vibration Associated with Disturbance from Road Wheels,” SAE interests are vehicular communication, intelligent transportation
World Congress Exhibition, Detroit, MI, USA, 2006. systems, and autonomous driving systems.
[6] S.Y. Cho and H.K. Lee, “Modified RHKF Filter for Improved
DR/GPS Navigation against Uncertain Model Dynamics,” ETRI Wooyong Han received his BS and MS
J., vol. 34, no. 3, June 2012, pp. 379–387. degrees in electronics engineering from the
[7] M. Enkhtur, S.Y. Cho, and K.-H. Kim, “Modified Unscented Engineering College, Kyung Hee University,
Kalman Filter for a Multirate INS/GPS Integrated Navigation Yongin, Rep. of Korea, in 1983 and 1985 and
System,” ETRI J., vol. 35, no. 5, Oct. 2013, pp. 943–946. his PhD degree in computer science from
[8] B. Park and W.Y. Han, “Reference Velocity Estimation Method the Department of Computer Science and
for O-Rad Unmanned Ground Vehicle,” Int. Conf. Ubiquitous Engineering, Chungnam National University,
Robots Ambient Intell., Jeju, Rep. of Korea, Oct. 30–Nov. 2, 2013, Daejeon, Rep. of Korea, in 2005. Since 1989, he has been with the
pp. 230–231. Electronics and Telecommunications Research Institute, Daejeon, Rep.
[9] C. Chen and H.-S. Tan, “Experimental Study of Dynamic Look- of Korea, where he is now a principal researcher. His main research
ahead Scheme for Vehicle Steering Control,” American Contr. interests are object-oriented distributed computing; XML-based e-
Conf., San Diego, CA, USA, June 2–4, 1999, pp. 3163–3167. commerce common frameworks; XML-based B2B system
[10] S.F. Campbell, “Steering Control of an Autonomous Ground interoperability; XML digital signature and encryption; frameworks for
Vehicle with Application to the DARPAR Urban Challenge,” M.S. telematics applications independent of mobile networks; vehicles; and
thesis, Dept. Mechanical Eng., Massachusetts Institute of territorial digital multimedia broadcasting.

ETRI Journal, Volume 37, Number 3, June 2015 Myungwook Park et al. 625
http://dx.doi.org/10.4218/etrij.15.0114.0123