The Control of An Automatic Door Using Fuzzy

Logic
Harun SÜMBÜL Akif COùKUN Muharrem Taúdemir
Gumushane University TRT Genel Müdürlü÷ü Gumushane University
Electrical-Electronics Engineering Ankara, TURKEY Mechanical Engineering
Baglarbasi, Gumushane, TURKEY [email protected] Baglarbasi, Gumushane, TURKEY
[email protected] [email protected]

Abstract—in this paper; a fuzzy logic have designed to the control Prenciple; In classical systems, the system is checked
of an automatic door. Input status (WS: Walking Speed and DD: electronic sensor by walking toward the door at a moderate
Distance Door) and the output status (DOS: Door Opening speed. When about 5’ (1524mm) from the opening , it should
Speed) is determined. According to these statuses, rule bases (25 smoothly start sliding open and then stop without impact.
rules) have obtained. And the rules bases have been processed by
Repeat on the other side of the opening. Move slowly through
a fuzzy logic and have tried to control of an automatic door. As a
result; in this study, we have obtain very good results in control the door at 6” (152mm) per sec. The door should remain open
of an automatic door with fuzzy logic. The results of analyses (See Figure 1) [7].
have indicated that the controls performed with fuzzy logic
provided both less time, less probability and reliable, consistent
controls and that are feasible to in real.

Keywords- Automatic Door; Fuzzy Logic; Rule Bases; Control.

I. INTRODUCTION
Fuzzy logic controllers will be used in solving the problems
outlined in the previous paragraphs. Since the time that Lotfi Figure 1. Automatic Door [1]
A. Zadeh proposed a new approach to the analysis of complex
systems and decision processes in 1973 [1], fuzzy logic has
been applied in many areas of science and engineering.
However, fuzzy logic actually gained popularity when it was II. DESIGN OF A FUZZY LOGIC CONTROLLER
applied to industrial problems by engineers [2]. This popularity
can be attributed to the fact that fuzzy logic provides a
powerful vehicle that allows engineers to incorporate human Fuzzy logic controller is one of the recent developing
reasoning in the control algorithm. In contrast with the methods in control that earned its popularities. The idea
conventional control theory, fuzzy logic design is not solely behind the fuzzy logic controller is to write the rules that
based on a mathematical model of the plant; although having operating the controller in heuristic manner, mainly in If A
an approximate mathematical model helps in the fine tuning of Then B format. In general, as shown in Fig. 2, fuzzy logic
a fuzzy logic controller. A fuzzy logic controller implements controller is constructed by the following elements [8]:
human reasoning that has been programmed into fuzzy logic
language (fuzzification, membership functions, rules and
• A rule base (a set of If-Then rules), which contains
defuzzification) for controlling a plant [3]-[5].
a fuzzy logic quantification of the
There are a lot of methods relevant to control of automatic expert’s linguistic description of how to achieve good control.
doors, for instance; PIR sensors, Microcontroller, PID • An inference mechanism (also called an “inference
controller. However, our used method that Fuzzy Logic is engine” or “fuzzy inference” module),
considerably a new method to the control of automatic doors, which emulates the expert’s decision making in interpreting
including input-output variables, rule bases, fuzzification, and applying knowledge about
defuzzification. how best to control the plant.
Automatic doors exemplify the challenges of designing • A fuzzification interface, which converts controller
emotionally welcoming interactive systems—a critical issue in input into information that the inference
the design of any system of incidental use [6]. mechanism can easily being used to activate and apply rules.
• A defuzzification interface, which converts the
Below, there are all of these descriptions and principles
conclusions of the inference mechanism into
about this subject.
actual inputs for the process.

978-1-61284-922-5/11/$26.00 ©2011 IEEE

432
 B. Membership functions:

1. For WS:

­10 − x , 0 ≤ x ≤ 10
VS; μVS(x) = ° (1)
® 10
°̄ 0 , x ≥ 10 or x < 0
Figure 2. Fuzzy Controller Structure [8].

The main features of the fuzzy logic design process consist of

the development of input and output of the membership ­ x
functions [9]. the basic triangle and trapezoidal forms are ° 10 , 0 ≤ x ≤ 10
chosen for input and output membership functions. In most °° 20 − x
cases, the performance of fuzzy control is minimally S; μS(x) =
® , 10 ≤ x ≤ 20 (2)
influenced by the shapes of memberships, but mainly by ° 10 , x ≥ 20 or x < 0
the characteristics of control rules[10]. ° 0
°¯
In the design process of the Fuzzy Logic Controller, fuzzy
rules are defined for two input variables, and single output ­ x − 10
variable. Figure 3 shows this input end output variables (input ° 10 , 10 ≤ x ≤ 20
= WS and DD, output = DOS). °° 30 − x
M; μM(x) =
, 20 ≤ x ≤ 30 (3)
®
° 10 , x ≥ 30 or x ≤ 10
° 0
¯°

­ x − 25
° 10 , 25 ≤ x ≤ 35
°° 45 − x
F; μF(x) = ® , 35 ≤ x ≤ 45 (4)
° 10
, x ≤ 25 or x ≥ 45
Figure 3. FIS Editor
° 0
°¯
A. Input and Output Variables
­ 0 , x ≤ 40
1. Input Variables: °
VF; μVF(x) = ° (5)
WS: Person’s Walking Speed [0-50 m/min]. ® x − 40
° 10 , 40 ≤ x ≤ 50
WS= VS (Very Slow) – S (Slow) – M (Medium) – F(Fast) – VF(Very Fast) °
¯
DD: The Distance to Door of person [0-6 m].
DD= VC (Very Close) – C (Close) – M (Medium) – A (Away) – TA (Too
2. For DD:
Away)

2. Output Variables: ­1 − y , 0 ≤ y ≤ 1
VC; μVC(y) = ® (6)
DOS: The Door Opening Speed [0-500 rpm]. ¯ 0 , y ≥ 1 or x < 0
WS= VS (Very Slow) – S (Slow) – M (Medium) – F(Fast) – VF(Very Fast)
­ y , 0 ≤ y ≤1
°
C; μC(y) = ®2 − y , 1 ≤ y ≤ 2 (7)
° 0 , y≥2
¯

433
 ­ y − 1,5 , 1,5 ≤ y ≤ 2,5 The decision which the fuzzy controller makes is derived
°
M; μM(y) = ®3,5 − y , 2,5 ≤ y ≤ 3,5 (8) from the rules which are stored in the database. These are
° 0 , y ≥ 3,5 or y ≤ 1,5 stored in a set of rules. Basically the rules are if-then
¯ statements that are intuitive and easy to understand, since they
are nothing but common English statements. Rules used in this
­ y − 2,5 , 2,5 ≤ y ≤ 3,5 paper are derived from common sense, data taken from typical
F; μF(y) = ° (9)
®4,5 − y , 3,5 ≤ y ≤ 4,5 use, and experimentation in a controlled environment.
° 0 , y ≥ 4,5 or y ≤ 2,5
¯ The sets of rules used here to derive the output are:

­ 0 , y≤5 TABLE I. THE SETS OF RULES

VF; μVF(y) = ° (10)
®y − 5 , 4 ≤ y ≤ 6 Rule 1: If WS is VS And DD is TA Then DOS is VS.
° Rule 2: If WS is VS And DD is A Then DOS is VS.
¯
Rule 3: If WS is VS And DD is M Then DOS is S.
Rule 4: If WS is VS And DD is C Then DOS is S.
3. For DOS: Rule 5: If WS is VS And DD is VC Then DOS is M.
Rule 6: If WS is S And DD is TA Then DOS is VS.
Rule 7: If WS is S And DD is A Then DOS is S.
Rule 8: If WS is S And DD is M Then DOS is S.
­
°100 − z Rule 9: If WS is S And DD is C Then DOS is M.
VS; μVS (z) =
® , 0 ≤ z ≤ 100 (11) Rule 10: If WS is S And DD is VC Then DOS is F.
° 100 , z ≥ 100 Rule 11: If WS is M And DD is TA Then DOS is S.
¯ 0 Rule 12: If WS is M And DD is A Then DOS is S.
Rule 13: If WS is M And DD is M Then DOS is M.
Rule 14: If WS is M And DD is C Then DOS is F.
­ z − 50 Rule 15: If WS is M And DD is VC Then DOS is F.
° 100 , 50 ≤ z ≤ 150
°° 250 − x Rule 16: If WS is F And DD is TA Then DOS is S.
S; μS (z) =
, 150 ≤ z ≤ 250 (12) Rule 17: If WS is F And DD is A Then DOS is M.
®
° 100 , z ≥ 250 or z ≤ 50
Rule 18: If WS is F And DD is M Then DOS is F.
° 0 Rule 19: If WS is F And DD is C Then DOS is F.
°¯ Rule 20: If WS is F And DD is VC Then DOS is VF.
Rule 21: If WS is VF And DD is TA Then DOS is M.
Rule 22: If WS is VF And DD is A Then DOS is F.
­ z − 200 Rule 23: If WS is VF And DD is M Then DOS is F.
° 100 , 200 ≤ z ≤ 300 Rule 24: If WS is VF And DD is C Then DOS is VF.
M; μM (z) = °° 400 − z (13)
, 300 ≤ z ≤ 400 Rule 25: If WS is VF And DD is VC Then DOS is VF.
®
° 100 , z ≥ 400 or z ≤ 200
° 0
°¯
TABLE II. RULE TABLE FOR PROBLEM

­ z − 350 DD
° 100 , 350 ≤ z ≤ 450 TA A M C VC
WS
F; μF (z) = °
° 550 − z (14) VS VS VS S S M
® , 450 ≤ z ≤ 550
° 100 , z ≥ 550 or z ≤ 350 S VS S S M F
° 0 M S S M F F
°¯ DOS
F S M F F VF
VF M F F VF VF
­ 0 , 0 ≤ z ≤ 400
° z − 400
VF; μVF (z) = ® , 400 ≤ z ≤ 500 (15)
These rules have been shown as membership functions in
° 100 figure 4, 5 and 6.
¯

434
 Figure 4. Membership function for DD

Figure 8. Rule viewer

Figure 5. Membership function for WS

IV. CONCLUSION
In this paper, an application of fuzzy algorithms in the
design of the fuzzy sets for the fuzzy logic controller of the
automatic door has been investigated.
By the use of fuzzy logic control we have been able to
obtain the aoutomatic door opening speed (0-500 rpm) for
different speed of walking (0-50 m/min) and different distance
of door (0-6 m). The conventional method required the human
interruption to decide upon what should be opening speed for
Figure 6. Membership function of DOS different variables. In other words this situation analysis
ability has been incorporated in the door which makes the
door much more automatic and represents the decision taking
III. SYSTEM PERFORMANCE AND TEST RESULTS
power of the new arrangement. Though the analysis in this
The sensors sense the input variables and using the above paper has been very crude, but this clearly depicts the
model the inputs are fuzzyfied and then by using simple if-else advantage of adding the fuzzy logic controller in the
rules and other simple fuzzy set operations the output fuzzy conventional aoutomatic door.
function is obtained and using the criteria the output variable
for DOS (Door Opening Speed) is obtained. Figure 5 shows
the response surface of the input-output relations as REFERENCES
determined by FIU. FIU stands for Fuzzy Interface Unit. This
is the fundamental unit in which the application interface [1] Zadeh, L. A.: ‘Outline of a new approach to the analysis of complex
FIDE encodes controller information. systems and decision processes’, IEEE Trans. on Systems, Man and
Cybernetics, vol. SMC-3, no. 1, pp. 28-44, 1973.
[2] Mamdani, E. H. and Assilian, S.: ‘An experiment in linguistic synthesis
with a fuzzy logic controller’, International Journal of Man-Machine
Studies, vol. 7, 1975, pp 1-13.
[3] Yager, R. R. and Filev, D. P.: ‘Essentials of fuzzy modelling and
control’, John Wiley&Sons, 1994.
[4] Mendel, J., M., “Fuzzy logic systems for engineering: a tutorial”,
Proceedings of the IEEE, vol. 83, no. 3, March 1995, pp. 345-377.
[5] Tsoukalas, L. H. and Uhrig, R. E., “Fuzzy and neural approaches in
engineering”, 1997.
[6] Ju, W., & Takayama, L., Approachability: How people interpret
Figure 7. Input/output response surface
automatic door movement as gesture. International Journal of Design,
3(2), 2009.
The results (the above plot) shows the way the automatic door [7] G. Horton Automatic, 4242 Baldwin Boulevard Corpus Christi, Texas,
http://www.hortondoors.com/.
will response in different conditions. For example, if we take
[8] K.M. Passino and S. Yurkovich, Fuzzy Control, Addison Wesley
WS= 27.8 and DD= 3.46, DOS which the model output is Longman, 1998.
equivalent to 240 RPM. This is quite convincing and [9] Wahyudi and J. Jalani, Proceedings of the 2nd International Conference
appropriate. on Mechatronics, ICOM'05, 10-12 May 2005, Kuala Lumpur, Malaysia.
[10] H.H. Lee and S.K. Cho, “A New Fuzzy Logic Anti_Swing Control for
Industrial Dimensional, Overhead Cranes”, Thesis, Department of
Mechanical Engineering, University of Suwon.

435