We consider a class of robust optimization problems that we call “robust-to-dynamics optimization” (RDO). The input to an RDO problem is twofold: (i) a mathematical program (e.g., an LP, SDP, IP, etc.), and (ii) a dynamical system (e.g., a linear, nonlinear, discrete, or continuous dynamics). The objective is to maximize over the set of initial conditions that forever remain feasible under the dynamics. The focus of this paper is on the case where the optimization problem is a linear program and the dynamics are linear. We establish some structural properties of the feasible set and prove that if the linear system is asymptotically stable, then the RDO problem can be solved in polynomial time. We also outline a semidefinite programming based algorithm for providing upper bounds on robust-to-dynamics linear programs.
[1]
Alberto Bemporad,et al.
Robust model predictive control: A survey
,
1998,
Robustness in Identification and Control.
[2]
L. Khachiyan.
Polynomial algorithms in linear programming
,
1980
.
[3]
I. P. Kornfelʹd,et al.
Ergodic Theory
,
1963
.
[4]
Tamás Terlaky,et al.
A Survey of the S-Lemma
,
2007,
SIAM Rev..
[5]
C. Caramanis.
What is ergodic theory
,
1963
.
[6]
Constantine Caramanis,et al.
Theory and Applications of Robust Optimization
,
2010,
SIAM Rev..
[7]
Laurent El Ghaoui,et al.
Robust Optimization
,
2021,
ICORES.