Docstoc

蟻群算法應用在足球機器人的避障路徑上

Document Sample
蟻群算法應用在足球機器人的避障路徑上 Powered By Docstoc
					Southern Taiwan University

      GA-PSO Fuzzy Controller Design
      Method(1/11)
      The procedure of GA-PSO algorithm can be described as follows:

           Step 1: Initialize the PSO algorithm by setting
                   F1 pbest  F2pbest   FNpbest  0 , g  1 the maximum
                   number of generation (G), the number of particles (N),
                                                      c1 , c2 
                               and four parameter values ,of max    min
            and        .




                                           1
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(2/11)
            Step 2: Generate the initial position vector
                           h          
                          p1  p1h ,1 , p1h ,2 ,           , p1h , j  ,       , p1h ,n    
                    and the initial velocity vector
                           h          
                          v1  v1h ,1 , v1h ,2 ,      , v1h , j  ,        , v1h ,n    
                    of N particles randomly by
                                         j          
                          p1h , j   p min  p max  pmin rand ()
                                                 j        j                                         (12)

                    and

                          v1h, j    
                                        v   max
                                             j      v min 
                                                       j
                                                               rand ()
                                                   20                                                 (13)



                                                            2
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(3/11)
          Step 3: Calculate the fitness value of each particle in the g-th
                  generation by
                         F  ph   fit  ph  , h  1,2,
                              g            g
                                                            ,N
                                                    Pbest
          Step 4: Determine F         Pbest
                                               and ph for each particle by
                                      h

                                    Fhg , if FhPbest  Fhg
                                   
                         F Pbest
                                  Pbest
                          h
                                    Fh , otherwise               (15)
                                   
                         h  1, 2, , N 

                   and
                                      ph , if FhPbest  Fhg
                                     
                                        g

                         p Pbest
                           h         Pbest
                                      ph , otherwise
                                                                 (16)
                         h  1, 2,     , N
                                                     3
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(4/11)
          Step 5: Find an index q of the particle with the highest fitness by

                             q  arg max FhPbest
                                   h1,2, , N                    (17)
                  and determine F Gbest and PGbest by

                          F Gbest  FqPbest  max FhPbest          (18)
                                                   h1,2, , N 



                  and
                                   pGbest  pq
                                             Pbest                 (19)




                                                       4
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(5/11)
          Step 6: If g=G, then go to Step 12, Otherwise, go to Step 7.

          Step 7: Update the velocity vector of each particle by
                       vhg 1    vhg  c1  rand1()   pGbest  phg 
                                                                            (20)
                           c2  rand 2()   p   Pbest
                                                  h       p  g
                                                              h   
                   is a weight value and defined by
                                       max  min
                         max                          g
                                              G                             (21)




                                                          5
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(6/11)
          Step 8: With fixed-length chromosomes that the problem is
                  variable domains, select the number of chromosome
                  population is  , the crossover rate is pc , the mutation
                  rate is pm .

          Step 9: The definition of adaptive function to measure the problem
                  domain on a single chromosome of the performance or
                  adaptability. Adaptive function is built on the reproductive
                  process, the basis for selecting pairs of chromosomes.

          Step 10: The size of a randomly generated initial population of
                   chromosomes  .
                                 x1 , x2 ,   , x

                                               6
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(7/11)
          Step 11: Calculating the adaptability of each chromosomes.

                             f  x1  , f  x2  ,   , f  x 

          Step 12: In the current population, select a pair of chromosomes.
                   Parental chromosomes are selected and their adaptability
                   related to the rate. Adaptive chromosomes are selected
                   with high rate is higher than the low adaptability of the
                   chromosomes.

          Step 13: Through the implementation of genetic operators-crossover
                    and mutation of a pair of offspring chromosomes.



                                                      7
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(8/11)
          Step 14: The offspring chromosomes into new populations.

          Step 15: Repeat step 13, unit the new chromosome population size
                   is equal to the size of initial population  .

          Step 16: With the new (offspring) chromosome populations to
                   replace to the initial (parent) chromosome populations.

          Step 17: Back to step 12, repeat this process until you meet the
                    conditions for ending to stop.




                                           8
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(9/11)
          Step 18: Check the velocity constraint by

                                 v max , if v gh1j  v max
                                                    , 
                                 
                                    j                           j
                                 
                     vgh1j   vgh1j  , if v min  vgh1j   v max
                         ,            ,              j            ,    j     (22)
                                  min
                                 v j        if vgh1j   v min
                                                   ,         j

                     h  1,2, , N , j  1,2, , n


          Step 19: Update the position vector of each particle by

                        ph 1  ph  vh 1
                         g       g    g
                                                                             (23)



                                                         9
Southern Taiwan University

     GA-PSO Fuzzy Controller Design
     Method(10/11)
          Step 20: Bound the updated position vector of each particle in the
                   searching range by

                                       p max ,         if   pgh, 1j   p max
                                      
                                          j                                  j
                                      
                        pgh, 1j    pgh, 1j  ,   if   p min  pgh, 1j   p max
                                                               j                     j
                                       min                    g 1
                                                                                           (24)
                                       pj ,            if   p           p   min
                                                              h, j         j

                        h  1,2, , N ,                  j  1,2, , n



          Step 21: Let g=g+1 and go to Step 3.



                                                                 10
Southern Taiwan University

      GA-PSO Fuzzy Controller Design
      Method(11/11)
          Step 22: Determine the corresponding fuzzy controller based on
                                                  Gbest
                   the position of the particle p       with the best fitness
                   value F Gbest .

     In the above reasoning, we will use the FIRA simulator to confirm the
      results of our reasoning.




                                          11
Southern Taiwan University




      Simulation Results(1/2)
     The membership functions of d , a, y1 and y2 , as determined by the
      proposed GA-PSO based method, are presented in figure 4.




           (a)                       (b)                        (c)

      Figure 4. membership functions of (a) e1 and e2 (b) e3 and e4
      (c) y1 and y2 , as determined by the proposed GA-PSO method

                                       12
Southern Taiwan University




      Simulation Results(2/2)
     The figure 5 are the orbit of the soccer robot when controller by the
      proposed GA-PSO method and simulation with a FIRA simulation.




                     Figure5. The soccer robot moving

                                        13
Southern Taiwan University




      Conclusions
     The final results showed that, although the GA-PSO's convergence
      time is not the time than the PSO-based fast, and as we join the GA
      algorithm , after the results obtained will be closer to its optimal
      solution. In the future, we need to explore ways to let a faster
      convergence time for change




                                       14
Southern Taiwan University




           Thanks for your attention !




                             15

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:24
posted:12/4/2011
language:English
pages:15
liamei12345 liamei12345 http://
About