SB02PD

Solution of continuous-time algebraic Riccati equations (matrix sign function method) with error bounds and condition estimates

[Specification] [Arguments] [Method] [References] [Comments] [Example]

Purpose

  To solve the real continuous-time matrix algebraic Riccati
  equation

     op(A)'*X + X*op(A) + Q - X*G*X = 0,

  where op(A) = A or A' = A**T and G, Q are symmetric (G = G**T,
  Q = Q**T). The matrices A, G and Q are N-by-N and the solution X
  is an N-by-N symmetric matrix.

  An error bound on the solution and a condition estimate are also
  optionally provided.

  It is assumed that the matrices A, G and Q are such that the
  corresponding Hamiltonian matrix has N eigenvalues with negative
  real parts.

Specification
      SUBROUTINE SB02PD( JOB, TRANA, UPLO, N, A, LDA, G, LDG, Q, LDQ, X,
     $                   LDX, RCOND, FERR, WR, WI, IWORK, DWORK, LDWORK,
     $                   INFO )
C     .. Scalar Arguments ..
      CHARACTER          JOB, TRANA, UPLO
      INTEGER            INFO, LDA, LDG, LDQ, LDWORK, LDX, N
      DOUBLE PRECISION   FERR, RCOND
C     .. Array Arguments ..
      INTEGER            IWORK( * )
      DOUBLE PRECISION   A( LDA, * ), DWORK( * ), G( LDG, * ),
     $                   Q( LDQ, * ), WI( * ), WR( * ), X( LDX, * )

Arguments

Mode Parameters

  JOB     CHARACTER*1
          Specifies the computation to be performed, as follows:
          = 'X':  Compute the solution only;
          = 'A':  Compute all: the solution, reciprocal condition
                  number, and the error bound.

  TRANA   CHARACTER*1
          Specifies the option op(A):
          = 'N':  op(A) = A    (No transpose);
          = 'T':  op(A) = A**T (Transpose);
          = 'C':  op(A) = A**T (Conjugate transpose = Transpose).

  UPLO    CHARACTER*1
          Specifies which triangle of the matrices G and Q is
          stored, as follows:
          = 'U':  Upper triangles of G and Q are stored;
          = 'L':  Lower triangles of G and Q are stored.

Input/Output Parameters
  N       (input) INTEGER
          The order of the matrices A, G, Q, and X.  N >= 0.

  A       (input) DOUBLE PRECISION array, dimension (LDA,N)
          The leading N-by-N part of this array must contain the
          coefficient matrix A of the equation.

  LDA     INTEGER
          The leading dimension of the array A.  LDA >= max(1,N).

  G       (input) DOUBLE PRECISION array, dimension (LDG,N)
          If UPLO = 'U', the leading N-by-N upper triangular part of
          this array must contain the upper triangular part of the
          matrix G.
          If UPLO = 'L', the leading N-by-N lower triangular part of
          this array must contain the lower triangular part of the
          matrix G.

  LDG     INTEGER
          The leading dimension of the array G.  LDG >= max(1,N).

  Q       (input) DOUBLE PRECISION array, dimension (LDQ,N)
          If UPLO = 'U', the leading N-by-N upper triangular part of
          this array must contain the upper triangular part of the
          matrix Q.
          If UPLO = 'L', the leading N-by-N lower triangular part of
          this array must contain the lower triangular part of the
          matrix Q.

  LDQ     INTEGER
          The leading dimension of the array Q.  LDQ >= max(1,N).

  X       (output) DOUBLE PRECISION array, dimension (LDX,N)
          If INFO = 0, INFO = 2, or INFO = 4, the leading N-by-N
          part of this array contains the symmetric solution matrix
          X of the algebraic Riccati equation.

  LDX     INTEGER
          The leading dimension of the array X.  LDX >= max(1,N).

  RCOND   (output) DOUBLE PRECISION
          If JOB = 'A', the estimate of the reciprocal condition
          number of the Riccati equation.

  FERR    (output) DOUBLE PRECISION
          If JOB = 'A', the estimated forward error bound for the
          solution X. If XTRUE is the true solution, FERR bounds the
          magnitude of the largest entry in (X - XTRUE) divided by
          the magnitude of the largest entry in X.

  WR      (output) DOUBLE PRECISION array, dimension (N)
  WI      (output) DOUBLE PRECISION array, dimension (N)
          If JOB = 'A' and TRANA = 'N', WR and WI contain the real
          and imaginary parts, respectively, of the eigenvalues of
          the matrix A - G*X, i.e., the closed-loop system poles.
          If JOB = 'A' and TRANA = 'T' or 'C', WR and WI contain the
          real and imaginary parts, respectively, of the eigenvalues
          of the matrix A - X*G, i.e., the closed-loop system poles.
          If JOB = 'X', these arrays are not referenced.

Workspace
  IWORK   INTEGER array, dimension (LIWORK), where
          LIWORK >= 2*N,          if JOB = 'X';
          LIWORK >= max(2*N,N*N), if JOB = 'A'.

  DWORK   DOUBLE PRECISION array, dimension (LDWORK)
          On exit, if INFO = 0 or INFO = 2, DWORK(1) contains the
          optimal value of LDWORK. If JOB = 'A', then DWORK(2:N*N+1)
          and DWORK(N*N+2:2*N*N+1) contain a real Schur form of the
          closed-loop system matrix, Ac = A - G*X (if TRANA = 'N')
          or Ac = A - X*G (if TRANA = 'T' or 'C'), and the
          orthogonal matrix which reduced Ac to real Schur form,
          respectively.

  LDWORK  INTEGER
          The dimension of the array DWORK.
          LDWORK >= 4*N*N + 8*N + 1,               if JOB = 'X';
          LDWORK >= max( 4*N*N + 8*N + 1, 6*N*N ), if JOB = 'A'.
          For good performance, LDWORK should be larger, e.g.,
          LDWORK >= 4*N*N + 6*N +( 2*N+1 )*NB,     if JOB = 'X',
          where NB is the optimal blocksize.

          If LDWORK = -1, then a workspace query is assumed;
          the routine only calculates the optimal size of the
          DWORK array, returns this value as the first entry of
          the DWORK array, and no error message related to LDWORK
          is issued by XERBLA.

Error Indicator
  INFO    INTEGER
          = 0:  successful exit;
          < 0:  if INFO = -i, the i-th argument had an illegal
                value;
          = 1:  the Hamiltonian matrix has eigenvalues on the
                imaginary axis, so the solution and error bounds
                could not be computed;
          = 2:  the iteration for the matrix sign function failed to
                converge after 50 iterations, but an approximate
                solution and error bounds (if JOB = 'A') have been
                computed;
          = 3:  the system of linear equations for the solution is
                singular to working precision, so the solution and
                error bounds could not be computed;
          = 4:  the matrix A-G*X (or A-X*G) cannot be reduced to
                Schur canonical form and condition number estimate
                and forward error estimate have not been computed.

Method
  The Riccati equation is solved by the matrix sign function
  approach [1], [2], implementing a scaling which enhances the
  numerical stability [4].

References
  [1] Bai, Z., Demmel, J., Dongarra, J., Petitet, A., Robinson, H.,
      and Stanley, K.
      The spectral decomposition of nonsymmetric matrices on
      distributed memory parallel computers.
      SIAM J. Sci. Comput., vol. 18, pp. 1446-1461, 1997.

  [2] Byers, R., He, C., and Mehrmann, V.
      The matrix sign function method and the computation of
      invariant subspaces.
      SIAM J. Matrix Anal. Appl., vol. 18, pp. 615-632, 1997.

  [3] Higham, N.J.
      Perturbation theory and backward error for AX-XB=C.
      BIT, vol. 33, pp. 124-136, 1993.

  [4] Petkov, P.Hr., Konstantinov, M.M., and Mehrmann, V.,
      DGRSVX and DMSRIC: Fortran 77 subroutines for solving
      continuous-time matrix algebraic Riccati equations with
      condition and accuracy estimates.
      Preprint SFB393/98-16, Fak. f. Mathematik, Technical
      University Chemnitz, May 1998.

Numerical Aspects
  The solution accuracy can be controlled by the output parameter
  FERR.

Further Comments
  The condition number of the Riccati equation is estimated as

  cond = ( norm(Theta)*norm(A) + norm(inv(Omega))*norm(Q) +
              norm(Pi)*norm(G) ) / norm(X),

  where Omega, Theta and Pi are linear operators defined by

  Omega(W) = op(Ac)'*W + W*op(Ac),
  Theta(W) = inv(Omega(op(W)'*X + X*op(W))),
     Pi(W) = inv(Omega(X*W*X)),

  and the matrix Ac (the closed-loop system matrix) is given by
     Ac = A - G*X, if TRANA = 'N', or
     Ac = A - X*G, if TRANA = 'T' or 'C'.

  The program estimates the quantities

  sep(op(Ac),-op(Ac)') = 1 / norm(inv(Omega)),

  norm(Theta) and norm(Pi) using 1-norm condition estimator.

  The forward error bound is estimated using a practical error bound
  similar to the one proposed in [3].

Example

Program Text

*     SB02PD EXAMPLE PROGRAM TEXT.
*     Copyright (c) 2002-2017 NICONET e.V.
*
*     .. Parameters ..
      INTEGER          NIN, NOUT
      PARAMETER        ( NIN = 5, NOUT = 6 )
      INTEGER          NMAX
      PARAMETER        ( NMAX = 20 )
      INTEGER          LDA, LDG, LDQ, LDX
      PARAMETER        ( LDA = NMAX, LDG = NMAX, LDQ = NMAX,
     $                   LDX = NMAX )
      INTEGER          LIWORK
      PARAMETER        ( LIWORK = MAX( 2*NMAX, NMAX*NMAX ) )
      INTEGER          LDWORK
      PARAMETER        ( LDWORK = MAX( 4*NMAX*NMAX + 8*NMAX,
     $                                 6*NMAX*NMAX ) + 1 )
*     .. Local Scalars ..
      DOUBLE PRECISION FERR, RCOND
      INTEGER          I, INFO, J, N
      CHARACTER        JOB, TRANA, UPLO
*     .. Local Arrays ..
      DOUBLE PRECISION A(LDA,NMAX), DWORK(LDWORK), G(LDG,NMAX),
     $                 Q(LDQ,NMAX), WI(NMAX), WR(NMAX),
     $                 X(LDX,NMAX)
      INTEGER          IWORK(LIWORK)
*     .. External Functions ..
      LOGICAL          LSAME
      EXTERNAL         LSAME
*     .. External Subroutines ..
      EXTERNAL         SB02PD
*     .. Intrinsic Functions ..
      INTRINSIC        MAX
*     .. Executable Statements ..
*
      WRITE ( NOUT, FMT = 99999 )
*     Skip the heading in the data file and read the data.
      READ ( NIN, FMT = '()' )
      READ ( NIN, FMT = * ) N, JOB, TRANA, UPLO
      IF ( N.LT.0 .OR. N.GT.NMAX ) THEN
         WRITE ( NOUT, FMT = 99995 ) N
      ELSE
         READ ( NIN, FMT = * ) ( ( A(I,J), J = 1,N ), I = 1,N )
         READ ( NIN, FMT = * ) ( ( Q(I,J), J = 1,N ), I = 1,N )
         READ ( NIN, FMT = * ) ( ( G(I,J), J = 1,N ), I = 1,N )
*        Find the solution matrix X.
         CALL SB02PD( JOB, TRANA, UPLO, N, A, LDA, G, LDG, Q, LDQ, X,
     $                LDX, RCOND, FERR, WR, WI, IWORK, DWORK, LDWORK,
     $                INFO )
*
         IF ( INFO.NE.0 ) THEN
            WRITE ( NOUT, FMT = 99998 ) INFO
         END IF
         IF ( INFO.EQ.0 .OR. INFO.EQ.2 .OR. INFO.EQ.4 ) THEN
             WRITE ( NOUT, FMT = 99997 )
             DO 20 I = 1, N
                WRITE ( NOUT, FMT = 99996 ) ( X(I,J), J = 1,N )
   20        CONTINUE
             IF ( LSAME( JOB, 'A' ) .AND. INFO.NE.4 ) THEN
                WRITE ( NOUT, FMT = 99994 ) RCOND
                WRITE ( NOUT, FMT = 99993 ) FERR
            END IF
         END IF
      END IF
      STOP
*
99999 FORMAT (' SB02PD EXAMPLE PROGRAM RESULTS',/1X)
99998 FORMAT (' INFO on exit from SB02PD = ',I2)
99997 FORMAT (' The solution matrix X is ')
99996 FORMAT (20(1X,F8.4))
99995 FORMAT (/' N is out of range.',/' N = ',I5)
99994 FORMAT (/' Estimated reciprocal condition number = ',F8.4)
99993 FORMAT (/' Estimated error bound = ',F20.16)
      END
Program Data
 SB02PD EXAMPLE PROGRAM DATA
   2     A     N     U 
   0.0   1.0
   0.0   0.0
   1.0   0.0
   0.0   2.0
   0.0   0.0
   0.0   1.0
Program Results
 SB02PD EXAMPLE PROGRAM RESULTS

 The solution matrix X is 
   2.0000   1.0000
   1.0000   2.0000

 Estimated reciprocal condition number =   0.1333

 Estimated error bound =   0.0000000000000063

Return to index