Automotive Engineering - Lane Detection sensors

1. A glimpse at the advanced
technology used to control the
car motion.

2.  Today’s modern automobiles have a variety
of sensors.
 But how important are they to the user?
1. They can identify and prevent possible issues
before they result in breakdowns which may
cause high expensive repairs.
2. They ensure that the vehicle is operating at its
most efficient.

5.  Function:
It warns the driver when the vehicle begins to
move out of its lane (unless a turn signal is on in
that direction).
 Importance:
Minimize accidents by addressing the main
causes of collisions: driver errors or distractions.

7.  There are three types of systems:
1. Lane Departure Warning (LDW):
Systems which warn the driver if the vehicle is
leaving its lane.
Types of warning:
 Sound warning.
 Vibrating sound.
 Visual warning.

8. 2. Lane Keeping System (LKS):
Systems which warn the driver and, if no
action is taken, automatically take steps
to ensure the vehicle stays in its lane.
3. Lane Centering Assist (LCA):
Keep the car centered in its current
lane, and ask the driver to take over in
challenging situations.

9.  How it works?
The methods for lane departure warning can
be classified into two categories:
1. Sensor-based methods.
2. Vision-based methods.

10.  Sensor-based methods:
Use devices such as radar, laser sensors, and
even global positioning systems (GPS) to detect
whether a vehicle departed a lane based on:
* The information of the vehicle ahead.
* Or the position calculated by GPS.
 Advantages:
Scanning distance (up to 100 m)
Their high reliability in dust, snow, and other
poor weather conditions.
 Disadvantages:
Poor accuracy to detect the lane positions.
Their information are unreliable inside a tunnel
or if no other vehicle is ahead.

12.  Vision-based method:
It detects the road lanes based on the
features of a camera image such as color
gradient, histogram, or edge.
** A color histogram is a representation of
the distribution of colors in an image.

13.  Vision-based solutions may be divided into:
1) Model-based methods:
Which create a mathematical model of the
road structure.
They use the geometric
coordinates of the
camera and the road as
input parameters and
depend on their
accuracy.

14.  Using this method, lane detection becomes a
problem of solving mathematical models.
The accuracy of the
detection depends
on the initial input
parameters of the
camera and the
shape of the road
which is represented
in the feature
points extracted
from the captured
image.

15. 2) Feature-based methods:
 Which can discriminate feature points of lane
markings from the non-lane areas by
characteristic features of the road, such as
color, gradient, or edge.

16.  This method is simple, but they require a
clear and strong color contrast of the lane
and good road conditions with little changes
in the surrounding environment.

17.  Actuators:
Once the sensor senses unintentionally out-
lane drifting, it sends warnings to the driver
through:
A sound message.
A vibration through the driver seat.
A visual warning behind the wheel.

18.  References:
1. Wikipedia
2. Researches made by:
Abdullhakam, Othman
Department of Electrical & Computer
Faculty of Engineering
International Islamic University Malaysia
3. Researches made by:
Toan Minh Hoang and Kang Ryoung Park
Multidisciplinary Digital Publishing Institute
Basel, Switzerland