A multigrid method for constrained optimal control problems

We consider the fast and efficient numerical solution of linear-quadratic optimal control problems with additional constraints on the control. Discretization of the first-order conditions leads to an indefinite linear system of saddle point type with additional complementarity conditions due to the control constraints. The complementarity conditions are treated by a primal-dual active set strategy that serves as outer iteration. At each iteration step, a KKT system has to be solved. Here, we develop a multigrid method for its fast solution. To this end, we use a smoother which is based on an inexact constraint preconditioner. We present numerical results which show that the proposed multigrid method possesses convergence rates of the same order as for the underlying (elliptic) PDE problem. Furthermore, when combined with a nested iteration, the solver is of optimal complexity and achieves the solution of the optimization problem at only a small multiple of the cost for the PDE solution.

[1]  Nicholas I. M. Gould,et al.  Constraint Preconditioning for Indefinite Linear Systems , 2000, SIAM J. Matrix Anal. Appl..

[2]  J. L. Lions,et al.  Control of Systems Governed by Parabolic Partial Differential Equations , 1971 .

[3]  D. Carr.  Rheinische Friedrich-Wilhelms-Universität Bonn , 1993 .

[4]  Todd Arbogast,et al.  Enhanced Cell-Centered Finite Differences for Elliptic Equations on General Geometry , 1998, SIAM J. Sci. Comput..

[5]  Gene H. Golub,et al.  Numerical solution of saddle point problems , 2005, Acta Numerica.

[6]  George Biros,et al.  Parallel Lagrange-Newton-Krylov-Schur Methods for PDE-Constrained Optimization. Part I: The Krylov-Schur Solver , 2005, SIAM J. Sci. Comput..

[7]  Alfio Borzì,et al.  An Algebraic Multigrid Method for a Class of Elliptic Differential Systems , 2003, SIAM J. Sci. Comput..

[8]  Michael Hinze,et al.  A Variational Discretization Concept in Control Constrained Optimization: The Linear-Quadratic Case , 2005, Comput. Optim. Appl..

[9]  J. Lions.  Optimal Control of Systems Governed by Partial Differential Equations , 1971 .

[10]  George Biros,et al.  Parallel Lagrange-Newton-Krylov-Schur Methods for PDE-Constrained Optimization. Part II: The Lagrange-Newton Solver and Its Application to Optimal Control of Steady Viscous Flows , 2005, SIAM J. Sci. Comput..

[11]  V. Schulz,et al.  Multigrid optimization in applications , 2000 .

[12]  Kazufumi Ito,et al.  The Primal-Dual Active Set Strategy as a Semismooth Newton Method , 2002, SIAM J. Optim..

[13]  L. Ridgway Scott.  Handbook of Numerical Analysis, Volume II. Finite Element Methods (Part I) (P. G. Ciarlet and J. L. Lions, eds.) , 1994, SIAM Rev..

[14]  P. Wesseling.  An Introduction to Multigrid Methods , 1992 .

[15]  Philippe G. Ciarlet,et al.  Handbook of Numerical Analysis , 1976 .

[16]  Sanjay Mehrotra,et al.  On the Implementation of a Primal-Dual Interior Point Method , 1992, SIAM J. Optim..

[17]  Stephen J. Wright,et al.  Numerical Optimization , 2018, Fundamental Statistical Inference.

[18]  A. Brandt.  Rigorous quantitative analysis of multigrid, I: constant coefficients two-level cycle with L 2 -norm , 1994 .

[19]  Stephen J. Wright.  Primal-Dual Interior-Point Methods , 1997, Other Titles in Applied Mathematics.

[20]  Arnd Rösch,et al.  Superconvergence Properties of Optimal Control Problems , 2004, SIAM J. Control. Optim..

[21]  A. Fursikov.  Optimal Control of Distributed Systems: Theory and Applications , 2000 .

[22]  Karl Kunisch,et al.  A Multigrid Scheme for Elliptic Constrained Optimal Control Problems , 2005, Comput. Optim. Appl..

[23]  Richard S. Falk,et al.  Approximation of a class of optimal control problems with order of convergence estimates , 1973 .

[24]  Karl Kunisch,et al.  A Comparison of a Moreau-Yosida-Based Active Set Strategy and Interior Point Methods for Constrained Optimal Control Problems , 2000, SIAM J. Optim..

[25]  K. Kunisch,et al.  Primal-Dual Strategy for Constrained Optimal Control Problems , 1999 .

[26]  Wolfgang Hackbusch,et al.  Multi-grid methods and applications , 1985, Springer series in computational mathematics.

[27]  P. Raviart,et al.  A mixed finite element method for 2-nd order elliptic problems , 1977 .

[28]  A. V. Fursikov,et al.  Optimal control of distributed systems , 1999 .

[29]  Wolfgang Hackbusch,et al.  Fast solution of elliptic control problems , 1980 .