Docstoc

A Fuzzy Approach to Prevent Headlight Glare

Document Sample
A Fuzzy Approach to Prevent Headlight Glare Powered By Docstoc
					                                                      (IJCSIS) International Journal of Computer Science and Information Security,
                                                      Vol. 9, No. 2, February 2011




      A Fuzzy Approach to Prevent Headlight Glare
Mrs.Niraimathi.S                                                       Dr.M.Arthanari
P.G.Department of computer applications                                Director
N.G.M College                                                          Bharathidasan School of computer applications
Pollachi-642001, TamilNadu, India                                      Ellispettai-638116, TamilNadu, India
niraisenthil@yahoo.com                                                 arthanarimsvc@gmail.com

                                                       Mr.M.Sivakumar
                                                 Doctoral Research Scholar
                                        Anna University, Coimbatore, TamilNadu, India
                                                        sivala@gmail.com




Abstract: This paper proposes a fuzzy rule based design                  Headlight glare is the main challenge, when driving at
approach to prevent the Headlight glare emitted by the               night to the drivers. During night the drivers are affected by
oncoming vehicles on the Highways. This gradually                    the dazzling high intensity headlights, which puts off their
reduces accidents on the Highways as the driver of the               vision and results in accidents. The blinding effect may be
oncoming vehicle is put on a comfortable zone which                  nearly total, if the lights have not been switched from high
might otherwise blind the oncoming driver’s visibility. In           beam, but even on low beam there is significant discomfort
the conventional vehicles the illumination is adjusted               and reduced visibility. This paper proposes a Fuzzy based
manually by the driver. This fuzzy based approach has                approach to reduce the headlight glare. The fuzzy sensor and
the fuzzy sensor and the controller embedded inside the              the fuzzy controllers embedded in the windshield during its
windshield or fit on to it, generates ambient illumination           lamination process or fit on to the windshield, gives a
to the oncoming driver, there by not ruining the vision of           solution to the headlight glare. The Sensor includes the
the driver during night. This setup has to be embedded               operation of checking the light source, if of over
on to all the vehicles, so that it prevents the happening of         tolerance/under tolerance. There by the controller converting
accidents. Fuzzy sensor and the controller makes use of              it in to low intensity if of high intensity and vice versa,
the fuzzy rules. The light intensity emitted by the                  providing ambient light source.
oncoming vehicle received by the fuzzy sensor, is                        The light intensity(I) measured in Volts and the
fuzzified using triangular membership function and                   distance(D) in metres are received by the fuzzy sensor. The
checked for the tolerance limit. If not of acceptable limit;         input parameters received by the fuzzy sensor are crisp input
the fuzzy sensor forwards it to the fuzzy controller which           values (Numerical value). These crisp sets are converted in to
converts the light intensity to an ambient light source              fuzzy sets using the process of fuzzification and are
thereby defuzzifying the output.                                     evaluated using the fuzzy rules. The output light intensity(OI)
                                                                     calculated using the fuzzy rules is checked for the tolerance
  Key words- Fuzzy logic; fuzzy rules; fuzzy sensor; fuzzy           limit by the fuzzy sensor. If beyond the tolerance limit, the
controllers; fuzzification; defuzzification; Headlight glare         fuzzy sensor defuzzifies using Centroid of Area and then
                                                                     sends it to the fuzzy controller which converts it to ambient
                         I. INTRODUCTION                             light source. The process of fuzzification and defuzzification
                                                                     is also repeated in the fuzzy controller.
   Around the world more than 1.2 million people lose their
life in Road Accidents, every year. 3 to 4 % of Gross                   In [1] a fuzzy controller that controls the brake rate of the
National Product is lost in Road Accidents. One person is            vehicle has been stated. The speed of the vehicle for which
killed in Road Accidents, every three minutes in the World.          the brake has to be applied and distance of the vehicle from
Total number of annual road accident deaths is more than the         the point at which it has to stop are passed as the input
total population of Mauritius.                                       parameters to a fuzzifier. The controller compares these
                                                                     inputs with the rule base and gives the desired output.




                                                               155                               http://sites.google.com/site/ijcsis/
                                                                                                 ISSN 1947-5500
                                                           (IJCSIS) International Journal of Computer Science and Information Security,
                                                           Vol. 9, No. 2, February 2011


   In [2] automatic fuzzy controller which controls the                   fuzzification of these parameters, linguistic variables are used
switching of headlight intensity of automobiles has been                  (Table I, II, III). The input Intensity(I) consists of 6 fuzzy
proposed..                                                                sets, Distance(D) has 10 fuzzy sets and the output parameter
   [3,5,6,8,9,11,12] gives basic understanding of Crisp set, its          ouput Intensity consists of 6 fuzzysets.
conversion to Fuzzy sets, concepts of fuzzy controller, and
the knowledge about Fuzzy Expert system. This paper gives                      TABLE I. LINGUISTIC VARIABLES FOR INPUT INTENSITY I(V) AND THEIR
the Methodology used in the fuzzy sensor, fuzzification of                                             NUMERICAL RANGE

input variables, rule evaluation and defuzzification in section
II, Implementation in section III and conclusion in section                     Linguistic value        Notation          Numerical
IV.                                                                                                                       range
                                                                                JustNoticeable          JN                0-3.50
                                                                                Noticeable              N                 3.00-6.50
                      II. METHODOLOGY                                           Satisfactory            S                 5.00-8.50
                                                                                JustAcceptable          JA                7.00-10.50
   The fuzzy sensor with its input parameters I(input                           Disturbing              D                 9.00-12.50
intensity), D(distance) and the output parameter OI(Sensor                      UnBearable              UB                11.00-14.50
output) is clearly shown in Fig. 1. The figures below indicate
the demonstrations derived using MATLAB.

                                                                                  TABLE II. LINGUISTIC VARIABLES FOR DISTANCE D(MTS) AND ITS
                                                                                                       NUMERICAL RANGE

                                                                               Linguistic value          Notation              Numerical
                                                                                                                               range
                                                                               VeryClose                 VC                    0-25
                                                                               Close                     CL                    12-50
                                                                               VeryNear                  VN                    37-75
                                                                               Near                      N                     62-100
                                                                               ModeratelyNear            MN                    87-125
                                                                               ModeratelyFar             MF                    110-150
                                                                               Far                       F                     135-175
               Fig. 1. The structure of the fuzzy sensor                       VeryFar                   VF                    160-200
                                                                               PrettyVeryFar             PVF                   185-225
A. Fuzzy inference process                                                     BoundaryZone              BZ                    210-250

   Fuzzy inference process defines a set of fuzzy “if – then
“rules. Most fuzzy logic based system uses rule bases to                       TABLE III. LINGUISTIC VARIABLES FOR SENSOR OUTPUT LIGHT SOURCE
represent the relation among the linguistic variables and to                                    OI(V) AND ITS NUMERICAL RANGE
derive actions from sensor input.
The fuzzy inference process is performed in four steps:                        Linguistic value          Notation              Numerical
                                                                                                                               range
1.   Fuzzification of the input variables.                                     JustNoticeable            JN                    0-3.50
2.   Defining Membership functions.                                            Noticeable                N                     3.00-6.50
3.   Rule evaluation.                                                          Satisfactory              S                     5.00-8.50
4.   Defuzzification.                                                          JustAcceptable            JA                    7.00-10.50
                                                                               Disturbing                D                     9.00-12.50
                                                                               UnBearable                UB                    11.00-14.50
    1)    Fuzzification: Fuzzification is the process of
converting the crisp input variables to fuzzy variables. It is
the mapping of the range of input to set membership values
of each fuzzy variable. The crisp values got for the input                        2) Defining Membership functions: After fuzzification
parameters I and D are converted in to fuzzy sets. For                         is done, the next process is to define the membership




                                                                    156                               http://sites.google.com/site/ijcsis/
                                                                                                      ISSN 1947-5500
                                                            (IJCSIS) International Journal of Computer Science and Information Security,
                                                            Vol. 9, No. 2, February 2011


functions in the fuzzy sets for the input and output
parameters. The Triangular membership function is used
for constructing the fuzzy sets. The membership
function of the input parameters is shown by the figures
(2-3). The membership function of the output parameter
                                                                                µS
is shown in figure.4. Fuzzy membership expressions for
the input Intensity(I) and Distance(D) is given by
(Eq.(1-2)).




                                                                                µJA




      Fig.2. The membership function of I(inputIntensity)                       µD




µJN


                                                                                µUB




µN                                                                                                                                 (1)




                                                                     157                               http://sites.google.com/site/ijcsis/
                                                                                                       ISSN 1947-5500
                                                       (IJCSIS) International Journal of Computer Science and Information Security,
                                                       Vol. 9, No. 2, February 2011




                                                                           µN




      Fig. 3. The membership function of D(Distance)

                                                                           µMN




µVC
                                                                           µMF




µCL
                                                                           µF




µVN
                                                                           µVF




                                                                158                               http://sites.google.com/site/ijcsis/
                                                                                                  ISSN 1947-5500
                                                                    (IJCSIS) International Journal of Computer Science and Information Security,
                                                                    Vol. 9, No. 2, February 2011




                                                                                        TABLE IV. RULE MATRIX REPRESENTATION

                                                                                   D/I         JN       N          S         JA          D            UB
    µPVF
                                                                                   BZ          JN       JN         JN        S           JA           D
                                                                                   PVF         JN       JN         N         S           D            D
                                                                                   VF          JN       JN         N         S           D            D
                                                                                   F           JN       N          S         S           D            D
                                                                                   MF          JN       N          S         JA          D            D
                                                                                   MN          N        N          S         JA          D            D
                                                                                   N           S        S          S         JA          D            UB
                                                                                   VN          S        S          S         D           D            UB
                                                                                   CL          JA       JA         D         D           UB           UB
     µBZ                                                                           VC          JA       D          D         UB          UB           UB


                                                                                      From the rule matrix we are able to arrive at the rule bases.
                                                                                   We have 10*6 rules for the fuzzy sensor. The rule base
                                                                                   consists of antecedent part and the consequent part.
                                                     (2)                           Antecedent part consists of input linguistic variables that may
                                                                                   be combined using AND operators. Consequent part contains
                                                                                   the output of the fuzzy rule. The figure below (Fig. 5) shows
                                                                                   the rule base for the sensor. In the figure below, the value of
                                                                                   I=13.9, D=250 and OI=7.25. This implies that the output
                                                                                   light intensity is moderate; the sensor judges it to be of the
                                                                                   acceptable limit and it need not send it to the controller.




    Fig. 4. The membership function of OI(Output Intensity of sensor )



  3) Rule evaluation: The fuzzy input values are processed
using the set of rules. The rules in fuzzy control consist of a
condition, IF, followed by a control action, THEN. Each rule
processes the information using different input parameters;
the output of each rule is different. In order to construct the
fuzzy rules we construct rule matrix (Table IV) and rule
bases.                                                                                      Fig. 5. Computing the value of OI for I=13.9 and D=250

                                                                                      If the output light intensity I is higher(9.00 and above), the
                                                                                   sensor sends it to the controller and the fuzzy controller (Fig.
                                                                                   6) converts it into an ambient light source.




                                                                             159                               http://sites.google.com/site/ijcsis/
                                                                                                               ISSN 1947-5500
                                                                    (IJCSIS) International Journal of Computer Science and Information Security,
                                                                    Vol. 9, No. 2, February 2011


                                                                                   COI=7.45 . This implies that the output light intensity is
                                                                                   moderate; the sensor judges it not to be of the acceptable
                                                                                   limit and it sends it to the controller.




               Fig. 6. The structure of the Fuzzy Controller

  The Fuzzy controller accepts the OutputIntensity(OI), if it
is of either Disturbing(D),or UNBEARABLE(UB)                 it
converts it to an ambient light source. The Fuzzy controller’s
ouput is the ControllerOutputIntensity(COI)(Table V)

TABLE V. THE LINGUISTIC VARIABLES FOR COI AND ITS NUMERICAL RANGE

 Linguistic value                  Notation           Numerical
                                                      range                        Fig. 8. Computing the value of COI for OI=9.5
 ReduceLightSource                 RLS                9.00-14.50
 AmbientLightSource                ALS                0-9.00
                                                                                     4) Defuzzification: The fuzzy sets are converted to crisp
                                                                                   values. Here the fuzzy sets represented by OI(Sensors output)
                                                                                   and ols(fuzzy controllers output) are converted to crisp
                                                                                   sets(Numerical values). Centre of Area method has been
                                                                                   used. General formula for COA is (Eq:3)
                                                                                   z* = ∫µc(z)zdz
                                                                                        ________
                                                                                        ∫µc(z)dz                                   (3)



Fig. 7. The membership function for the ControllerOutputIntensity(COI)                                      III. IMPLEMENTATION

  The rule bases for the fuzzy controller(Table VI) is as                          The MATLAB Fuzzy Logic Toolbox has been used to
shown below.                                                                       encode fuzzy sets, membership functions, fuzzy rules and to
                                                                                   perform inference process for both the fuzzy sensor and the
TABLE VI. THE RULE BASES FOR THE FUZZY CONTROLLER                                  fuzzy controller.

          Rule       OI            COI                                                                          IV. CONCLUSION
          1          JN            ALS
          2          N             ALS                                               This paper has proposed a fuzzy rule based approach to
          3          S             ALS
                                                                                   prevent the headlight glare which in turn minimizes the
          4          JA            ALS
                                                                                   Accidents. The fuzzy sensor and the controller uses the
          5          D             RLS
          6          UB            RLS                                             fuzzy rule bases to control the intensity of light. The
                                                                                   conventional controllers would not be very efficient in
                                                                                   controlling the headlight glare as there would be discrete
  The figure below (Fig. 8) shows the rule base for the                            values either high/low beam but the fuzzy controller has the
sensor. In the figure below, the value of OI=9.5 and that of                       continuous light intensities rather than high/low beam. The




                                                                             160                                http://sites.google.com/site/ijcsis/
                                                                                                                ISSN 1947-5500
                                                                       (IJCSIS) International Journal of Computer Science and Information Security,
                                                                       Vol. 9, No. 2, February 2011


fuzzy sensor and the controller has to be embedded inside the
windshield or fit on to it, during the windshield lamination
process. This fuzzy system comprising the sensor and the
controller reduces the headlight glare and therefore reduces
the accidents on the highways during the travel at night. This                                               AUTHORS PROFILE
fuzzy system would be of greater boon to the drivers as the
driving becomes comfortable without ruining the vision of
                                                                                                      Niraimathi.S (niraisenthil@hotmail.com) is a second year
the driver at both the ends. This setup comprising the fuzzy                                          Doctoral Research Scholar at Mother Teresa women’s
sensor and the fuzzy controller, has to be put up on all the                                          University. She holds M.Phil in Computer Science from
                                                                                                      the Bharathiar University. She had her Bachelor’s degree
vehicles in order to prevent the happening of accidents.                                              in Computer Science and Master’s degree in Computer
                                                                                                      Applications from the Bharathiar University, India. She has
                                                                                                      9+ years of experience in teaching. She is at present
                                                                                      working as an Assistant Professor in Nallamuthu Gounder Mahalingam
                                                                                      College, Pollachi, India. Her area of expertise includes Fuzzy systems,
                                                                                      OOAD and compiler design.
                            REFERENCES
                                                                                                        Dr. M. Arthanari holds a Ph.D. in Mathematics from
[1] Nikunja K. Swain, “A Survey of Application of Fuzzy Logic in                                       Madras University as well as Masters Degree in Computer
Intelligent Transportation Systems (ITS) and Rural ITS”- Southeast Con,                                Science from BITS, Pilani. He was the professor and Head
Proceedings of IEEE, 2006, pp 85-89.                                                                   of Computer Science and IT Department at Tejaa Shakthi
                                                                                                       Institute of Technology for Women, Coimbatore, India. At
[2] Kher.S, Bajaj .P, “Fuzzy control of head-light intensity of automobiles:                           present he is the Director, Bharathidhasan School of
design approach “proceedings of 37th SICE annual conference international                              Computer Applications, Ellispettai, Erode, Tamilnadu. He
session papers, July 1988, pp 1047-1050.                                              holds a patent issued by the Govt. of India for his invention in the field of
                                                                                      Computer Science. He has directed teams of Ph.D. researchers and industry
[3]George J. Klir, Bo Yuan, “Fuzzy Sets and Fuzzy logic theory and                    experts for developing patentable products. He teaches strategy, project
applications”, PHI Learning Private Ltd.                                              management, creative problem solving, innovation and integrated new
                                                                                      product development for last 36 years.
[4] T.M. Husain, T. Saadawi, S. Ahmed, “Overhead infrared sensor for
monitoring vehicular traffic,” IEEE Ttrans. On vehicular Tech, vol. 42, No. 4                           Sivakumar.M (sivala@gmail.com) has 10+ years of
pp.477-482, Nov 1993.
                                                                                                       experience in the software industry including Oracle
[5] R. Kruse, J. Gebhardt, F. Klawon, ”Foundations of Fuzzy Systems”,                                  Corporation. He received his Bachelor degree in Physics
Wiley, Chichester 1994.                                                                                and Masters in Computer Applications from the Bharathiar
                                                                                                       University, India. He holds a patent for the invention in
[6] Gerla G., Fuzzy Logic Programming and fuzzy control, Studia Logica,               embedded technology. He is technically certified by various professional
79 (2005) 231-254.                                                                    bodies like PRINCE2, ITIL, IBM Rational Clearcase Administrator, OCP -
                                                                                      Oracle Certified Professional 10G and ISTQB.
[7] Hájek P., Metamathematics of Fuzzy Logic, Kluwer Academic
Publishers, Dordrecht, The Netherlands, 1998.

[8] Lennart Ljung," An Introduction to Fuzzy Control”, NHI, 1992.

[9] F. Martin Mcheill, Thro, Yager," Fuzzy Logic- A Practlcal Approach',
A.P., 1994.

[10] Das, S. and Bowles, B.A., “Simulations of highway chaos using fuzzy
logic” 18th International Conference of the North American Fuzzy
Information Processing Society, 1999, pp. 130 -133.

[11] Zimmermann, H.J.: Fuzzy Set Theory And Its Applications. 2nd ed.,
Kluwer, 1990.

[12] Zadeh, L.A.: Fuzzy sets as a basis for a theory of possibility. Fuzzy
Sets and Systems. 1 (1978) pp. 3-28




                                                                                161                                 http://sites.google.com/site/ijcsis/
                                                                                                                    ISSN 1947-5500

				
DOCUMENT INFO
Description: The International Journal of Computer Science and Information Security (IJCSIS Vol. 9 No. 2) is a reputable venue for publishing novel ideas, state-of-the-art research results and fundamental advances in all aspects of computer science and information & communication security. IJCSIS is a peer reviewed international journal with a key objective to provide the academic and industrial community a medium for presenting original research and applications related to Computer Science and Information Security. . The core vision of IJCSIS is to disseminate new knowledge and technology for the benefit of everyone ranging from the academic and professional research communities to industry practitioners in a range of topics in computer science & engineering in general and information & communication security, mobile & wireless networking, and wireless communication systems. It also provides a venue for high-calibre researchers, PhD students and professionals to submit on-going research and developments in these areas. . IJCSIS invites authors to submit their original and unpublished work that communicates current research on information assurance and security regarding both the theoretical and methodological aspects, as well as various applications in solving real world information security problems.