Intelligent Traffic Lights
Control By Fuzzy Logic
Ranganath Muthu
Professor, EEE
SSN College of Engineering
19 December 2008 FLC Traffic 1
Fuzzy Logic Controller
• FLCs allow for a simpler human like approach
to control system design.
• They do not need the mathematical model of
the process.
• For non-linear systems, controlling with
conventional controllers is difficult.
• FLCs provide reasonable and effective
alternatives to classical controllers.
19 December 2008 FLC Traffic 2
Conventional Control System
19 December 2008 FLC Traffic 3
Fuzzy Logic Control System
19 December 2008 FLC Traffic 4
Traffic Lights
n Have you ever thought
that a green light should
have lasted a couple
seconds longer?
n Or been stuck at a red
light when there was no
traffic coming the other
way?
19 December 2008 FLC Traffic 5
n “Grrrrrr !!!”
If only traffic lights were more
intelligent…
19 December 2008 FLC Traffic 6
n Fuzzy logic technology
allows the
implementation of real-
life rules similar to the
way humans would
think. (Khiang)
19 December 2008 FLC Traffic 7
Conventional Traffic Lights
Control
n Preset cycle time
n A busy street may need a regularly timed
cycle of green lights.
n Preset cycle time and proximity sensors
n A less traveled street only needs a green
light when cars are present
n Cycle time and placement of sensors should
be customized for the particular road
19 December 2008 FLC Traffic 8
Fuzzy Traffic Lights Control
System
n Uses sensors that count cars instead of
proximity sensors, which only indicate
the presence of cars
n Provides the controller with traffic densities
in the lanes
n Allows a better assessment of changing
traffic patterns
19 December 2008 FLC Traffic 9
General Structure of Fuzzy Traffic
Lights Control System
19 December 2008 FLC Traffic 10
Design Criteria and
Constraints
n Four-way junction with traffic coming from the north,
west, south, and east
n When traffic from the north and south moves, traffic
from the east and west stops, and vice-versa
n No left and right turns are considered
n The fuzzy logic controller will observe north/south
traffic as one side and west/east traffic as another
side
n The west/east lane is the main approach
n Minimum green light time is 20 seconds and maximum
green light time is 2 minutes
19 December 2008 FLC Traffic 11
Design of FLC
Fuzzy Logic Control System
19 December 2008 FLC Traffic 12
Fuzzy Input Variables
n Arrival
n The quantity of traffic on the arrival side
n Note: if the north/south side is green, then
this would be the arrival side
n Queue
n The quantity of traffic on the queuing side
n Note: if the north/south side is not green,
then this would be the queuing side
19 December 2008 FLC Traffic 13
Fuzzy Output Variable
n Extension
n The extension time needed for the green light
on the arrival side
The value of this fuzzy output variable will
result in either extending or not extending
the current green light time
19 December 2008 FLC Traffic 14
Arrival Membership Functions
n AN = almost no
n F = few
n MY = many
n TMY = too
many
19 December 2008 FLC Traffic 15
Queue Membership Functions
n VS = very small
n S = small
n M = medium
n L = large
19 December 2008 FLC Traffic 16
Extension Membership Functions
n Z = zero
n S = short
n M = medium
n L = longer
19 December 2008 FLC Traffic 17
Fuzzy Inference
n IF Arrival is TMY
AN F MY TMY AND Queue is VS
THEN Extension is L
VS Z S M L n IF Arrival is F AND
S Z S M M Queue is VS THEN
Extension is S
M Z Z S M n IF Arrival is AN
L Z Z Z S AND Queue is VS
THEN Extension is Z
19 December 2008 FLC Traffic 18
Defuzzification
n The final output membership function for each
rule is the fuzzy set assigned to that output by
clipping the degree of truth values of the
membership functions of the associated
antecedents (Arrival and Queue).
n Once the membership degree of each output
fuzzy variable is determined, all of the rules that
are being fired are then combined and the
actual crisp output is obtained using the center
of gravity technique.
19 December 2008 FLC Traffic 19
Now You Try It!
Suppose three cars have arrived and five cars
are queued.
a) Write all fuzzy implication rules whose
antecedents have nonzero membership
values
If F and M, then Z
If F and L, then Z
If MY and M, then S
If MY and L, then Z
19 December 2008 FLC Traffic 20
Now You Try It!
b) Find the fuzzy output membership values for
the rules in (a) using the min operator.
If F = 0.5 and M = 0.5, then Z = 0.5
If F = 0.5 and L = 0.5, then Z = 0.5
If MY = 0.5 and M = 0.5, then S = 0.5
If MY = 0.5 and L = 0.5, then Z = 0.5
19 December 2008 FLC Traffic 21
Now You Try It!
c) Clip the final output membership function
using the values in (b).
19 December 2008 FLC Traffic 22
Now You Try It!
c) Clip the final output membership function
using the values in (b).
19 December 2008 FLC Traffic 23
Now You Try It!
d) Estimate the extension time by approximating
the center of gravity.
19 December 2008 FLC Traffic 24
Now You Try It!
d) Estimate the extension time by approximating
the center of gravity.
Approximately 1.0 seconds.
19 December 2008 FLC Traffic 25
Conclusion
n The fuzzy logic control system reacts the
way a traffic policeman would at a typical
intersection
n The fuzzy logic control system provides
better performance in terms of total
waiting time as well as total moving time,
reducing fuel consumption, air pollution,
and noise pollution
19 December 2008 FLC Traffic 26
References
§ Driankov D., Hellendoorn H. & Reinfrank
M., An Introduction to Fuzzy Control,
Narosa Publishing House New Delhi,
India, 1996.
§ Muthu R. & Ghosh S., Fuzzy Logic based
Traffic Lights Control System, Safety on
Roads International Conference,
Bahrain, 21-23 October 2002.
19 December 2008 FLC Traffic 27
Thank You
19 December 2008 FLC Traffic 28
