Page 910 - The Mechatronics Handbook
P. 910

0066_Frame_C30  Page 21  Thursday, January 10, 2002  4:44 PM








                       Since

                                                          D 12 C 1 =  0                        (30.105)
                                                           T
                                                          B 1 D T  =  0                        (30.106)
                                                             21
                       the imaginary axis rank conditions involving (A, B 2 , C l , D 12 ) and (A, B 1 , C 2 , D 21 ) in Assumption 30.2
                       become equivalent to (A,C l ) having no imaginary unobservable modes and (A, B 1 ) having no imaginary
                       uncontrollable modes. These are clearly satisfied since A = 1 has no imaginary modes. Given this, it
                                         2
                       follows that all of the H  output feedback problem assumptions in Assumption 30.2 are satisfied.
                         Plant. Finally, we note that the so-called plant (or missile) transfer function P = G 22  is given by

                                                                      −1
                                                               (
                                                                  –
                                                   P =  G 22 =  C 2 sI A) B 2                  (30.107)
                                                               1
                                                           =  ----------                       (30.108)
                                                              –
                                                             s 1
                       G 22  is unstable with a right half plane pole at s = 1. G 22  is also minimum phase (i.e., no zeros in Res > 0).
                                                  T
                         Filter Gain Matrix H f . Since B 1 D  =  0 , the associated FARE is given by
                                                  21
                                                          T
                                                                              ---Y =
                                                             1
                                                            –
                                       AY +  YA +  B 1 B 1 – YC 2 Θ C 2 Y =  Y + Y +  1 –  1  2  0  (30.109)
                                               T
                                                     T
                                                                              m
                       or
                                                        2
                                                               –
                                                       Y –  2mY m =  0                         (30.110)
                       Application of the quadratic formula and selecting the positive (stabilizing) root yields:
                                                                 2
                                                       Y =  m +  m +  m                        (30.111)
                       This yields the following filter gain matrix:

                                                                         1
                                                          T
                                                   H f =  YC 2 Θ – 1  =  1 +  1 +  ---         (30.112)
                                                                        µ
                       We now select m to achieve the given dominant pole specification:


                                                                      --- =
                                                 AH f C 2 =  11 –  1 +  1  – 5                 (30.113)
                                                   –
                                                             –
                                                                      m
                       This yields
                                                                1
                                                           m =  -----                          (30.114)
                                                               24
                       The associated KBF open loop transfer function is given by

                                                                    −1
                                                     G KF =  –  C 2 sI A) H f                  (30.115)
                                                              (
                                                                 –
                                                             6
                                                           –
                                                         =  ----------                         (30.116)
                                                            –
                                                           s 1
                       ©2002 CRC Press LLC
   905   906   907   908   909   910   911   912   913   914   915