Numerical solution to generalized Lyapunov/Stein and rational Riccati equations in stochastic control

We consider the numerical solution of the generalized Lyapunov and Stein equations in ℝn$\mathbb {R}^{n}$, arising respectively from stochastic optimal control in continuous- and discrete-time. Generalizing the Smith method, our algorithms converge quadratically and have an O(n3) computational complexity per iteration and an O(n2) memory requirement. For large-scale problems, when the relevant matrix operators are “sparse”, our algorithm for generalized Stein (or Lyapunov) equations may achieve the complexity and memory requirement of O(n) (or similar to that of the solution of the linear systems associated with the sparse matrix operators). These efficient algorithms can be applied to Newton’s method for the solution of the rational Riccati equations. This contrasts favourably with the naive Newton algorithms of O(n6) complexity or the slower modified Newton’s methods of O(n3) complexity. The convergence and error analysis will be considered and numerical examples provided.

[1]  T. Damm.  Rational Matrix Equations in Stochastic Control , 2004 .

[2]  Wen-Wei Lin,et al.  Large-scale Stein and Lyapunov equations, Smith method, and applications , 2013, Numerical Algorithms.

[3]  Peter Benner,et al.  A Semi-Discretized Heat Transfer Model for Optimal Cooling of Steel Profiles , 2005 .

[4]  V. Dragan,et al.  Mathematical Methods in Robust Control of Linear Stochastic Systems , 2006 .

[5]  G. Freiling,et al.  Basic properties of a class of rational matrix differential equations , 2001, 2001 European Control Conference (ECC).

[6]  H. Abou-Kandil,et al.  Matrix Riccati Equations in Control and Systems Theory , 2003, IEEE Transactions on Automatic Control.

[7]  Jens Saak,et al.  Efficient Numerical Solution of Large Scale Algebraic Matrix Equations in PDE Control and Model Order Reduction , 2009 .

[8]  Ivan Ganchev Ivanov,et al.  Iterations for solving a rational Riccati equation arising in stochastic control , 2007, Comput. Math. Appl..

[9]  Khalide Jbilou,et al.  Low rank approximate solutions to large Sylvester matrix equations , 2006, Appl. Math. Comput..

[10]  Chun-Hua Guo.  Iterative Methods for a Linearly Perturbed Algebraic Matrix Riccati Equation Arising in Stochastic Control , 2013 .

[11]  Gerhard Freiling,et al.  A survey of nonsymmetric Riccati equations , 2002 .

[12]  W. Wonham.  On a Matrix Riccati Equation of Stochastic Control , 1968 .

[13]  W. Wonham.  Erratum: On a Matrix Riccati Equation of Stochastic Control , 1969 .

[14]  Tobias Damm,et al.  Direct methods and ADI‐preconditioned Krylov subspace methods for generalized Lyapunov equations , 2008, Numer. Linear Algebra Appl..

[15]  D. Kressner,et al.  Greedy low-rank methods for solving general linear matrix equations ‡ , 2014 .

[16]  Ivan G. Ivanov,et al.  Properties of Stein (Lyapunov) iterations for solving a general Riccati equation , 2007 .

[17]  Daniel Kressner,et al.  Truncated low‐rank methods for solving general linear matrix equations , 2015, Numer. Linear Algebra Appl..

[18]  Wen-Wei Lin,et al.  Solving large-scale continuous-time algebraic Riccati equations by doubling , 2013, J. Comput. Appl. Math..

[19]  Gerhard Freiling,et al.  Properties of the solutions of rational matrix difference equations , 2003 .

[20]  Khalide Jbilou,et al.  An Arnoldi based algorithm for large algebraic Riccati equations , 2006, Appl. Math. Lett..

[21]  Chun-Hua Guo,et al.  Iterative Solution of a Matrix Riccati Equation Arising in Stochastic Control , 2002 .

[22]  Valeria Simoncini,et al.  A New Iterative Method for Solving Large-Scale Lyapunov Matrix Equations , 2007, SIAM J. Sci. Comput..

[23]  E. Chu,et al.  A modified Newton's method for rational Riccati equations arising in stochastic control , 2011, 2011 International Conference on Communications, Computing and Control Applications (CCCA).

[24]  I. Jaimoukha,et al.  Krylov subspace methods for solving large Lyapunov equations , 1994 .

[25]  Peter Benner,et al.  On the numerical solution of large-scale sparse discrete-time Riccati equations , 2011, Adv. Comput. Math..

[26]  P. Benner,et al.  Numerical solution of large and sparse continuous time algebraic matrix Riccati and Lyapunov equations: a state of the art survey , 2013 .

[27]  Valeria Simoncini,et al.  Computational Methods for Linear Matrix Equations , 2016, SIAM Rev..

[28]  Peter Benner,et al.  Low rank methods for a class of generalized Lyapunov equations and related issues , 2013, Numerische Mathematik.

[29]  Gene H. Golub,et al.  Matrix computations , 1983 .

[30]  Hung-Yuan Fan,et al.  Structure-Preserving Algorithms for Periodic Discrete-Time Algebraic Riccati Equations , 2004 .

[31]  Hung-Yuan Fan,et al.  A structure-preserving doubling algorithm for continuous-time algebraic Riccati equations , 2005 .

[32]  Bruno Iannazzo,et al.  Numerical Solution of Algebraic Riccati Equations , 2012, Fundamentals of algorithms.

[33]  V. Mehrmann.  The Autonomous Linear Quadratic Control Problem , 1991 .

[34]  Eric King-Wah Chu,et al.  Large-scale discrete-time algebraic Riccati equations - Doubling algorithm and error analysis , 2015, J. Comput. Appl. Math..

[35]  Peter Benner,et al.  Lyapunov Equations, Energy Functionals, and Model Order Reduction of Bilinear and Stochastic Systems , 2011, SIAM J. Control. Optim..

[36]  Wen-Wei Lin,et al.  Convergence Analysis of Structure-Preserving Doubling Algorithms for Riccati-Type Matrix Equations , 2006, SIAM J. Matrix Anal. Appl..

[37]  A. Laub,et al.  Benchmarks for the numerical solution of algebraic Riccati equations , 1997 .

[38]  Hung-Yuan Fan,et al.  Low-rank approximation to the solution of a nonsymmetric algebraic Riccati equation from transport theory , 2012, Appl. Math. Comput..

[39]  James Lam,et al.  On Smith-type iterative algorithms for the Stein matrix equation , 2009, Appl. Math. Lett..

[40]  H. Schneider.  Positive operators and an inertia theorem , 1965 .

[41]  Zhaojun Bai,et al.  A projection method for model reduction of bilinear dynamical systems , 2006 .

[42]  D. Hinrichsen,et al.  Newton's method for a rational matrix equation occurring in stochastic control , 2001 .

[43]  Khalide Jbilou,et al.  Block Krylov Subspace Methods for Large Algebraic Riccati Equations , 2003, Numerical Algorithms.

[44]  Leiba Rodman,et al.  Algebraic Riccati equations , 1995 .

[45]  Gerhard Freiling,et al.  On a class of rational matrix differential equations arising in stochastic control , 2004 .

[46]  M. Heyouni,et al.  AN EXTENDED BLOCK ARNOLDI ALGORITHM FOR LARGE-SCALE SOLUTIONS OF THE CONTINUOUS-TIME ALGEBRAIC RICCATI EQUATI ON ∗ , 2008 .

[47]  J. Brandts.  The Riccati algorithm for eigenvalues and invariant subspaces of matrices with inexpensive action , 2003 .