Page 243 - Mechatronics for Safety, Security and Dependability in a New Era
P. 243

Ch46-I044963.fm  Page 227  Tuesday, August 1, 2006  3:57 PM
                            Tuesday, August
                                           3:57 PM
                                      1, 2006
                      Page 227
            Ch46-I044963.fm
                                                                                          227
                                                                                          227
                  Filter Setup for  the Corners of Buildings
                  After  we  found  the  corresponding  corner  of building to the boundary  line Z = (x, d) T  in the  disparity
                  space, the observation equation can be described like the following  equation:
                                         = H(X,M)  + v
                                            m
                                         '  ( x -x p)s'm0 p  -(m y  -y p)cos0 p
                                           (m x-x p)cos0 p+(m y-y p)sm0 p                 (7)
                                                                    + v,
                                                     fl
                                            -
                                         (m x  x p ) cos 6 p  + (m y  -y p)sin0 p
                  where  M  = (m x,  m v) T  is  the  coordinates  of the  building  corner  on  the  map.  The  filter  setup  for  this
                  feature is as follows:
                                                   T         T                          (2)'"
                                            S = Jx^xJ x  + JMT.MJ M
                                                             M
                                                                                        (3)'"


                  EXPERIMENTAL RESULTS
                  The experimental results are shown in figure  3 magnified  from  figure 2 which shows the estimation  of
                  the robot pose by the localization  algorithms using the EKF. The color ellipses with  1 cr uncertainty  are
                  the estimated uncertainties of the robot poses by matching the features to the map.

                                                 Localization Results
                                    T [m]
                                      20
                                              B2       1
                                      15
                                      10
                                      5
                                      0
                                      -5
                                         :M3
                                     -10
                                     -15
                                     -20
                                     -25

                                       -30  -25  -20  -15  -10  -5  10  15  20
                                                      X [m]
                                   Figure 3: Localization  results with uncertainty ellipses.

                  Table  1 shows the  estimates  of the  robot pose  in each  feature  used  by the  localization  algorithm.  The
                  left  figures  of each table row represent the estimate of the localization method. The right parenthesized
                  figures  of the same row represent the standard deviation of the robot pose.
   238   239   240   241   242   243   244   245   246   247   248