Brief paperThe generalised discrete algebraic Riccati equation in linear-quadratic optimal control☆
Introduction
Due to their ubiquitousness in optimal control and filtering problems, as well as in linear factorisation and stochastic realisation problems, Riccati equations are universally regarded as a cornerstone of modern control theory. Several monographs have been entirely devoted to providing a general and systematic framework for the study of Riccati equations, see e.g. Abou-Kandil, Freiling, Ionescu, and Jank (2003), Ionescu, Oaraˇ, and Weiss (1999) and Lancaster and Rodman (1995).
The classic solution of the discrete-time infinite-horizon LQ problem is traditionally expressed in terms of the solution of the Riccati equation where and are such that the Popov matrix satisfies The set of matrices is often referred to as the Popov triple, see e.g. Ionescu et al. (1999). Eq. (1) is the so-called Discrete Riccati Algebraic Equation .
Nevertheless, an LQ problem may have solutions even if has no solutions, and the optimal control can be written in this case as a state feedback given in terms of a matrix such that is singular and satisfies the more general Riccati equation where the matrix inverse in has been replaced by the Moore–Penrose pseudo-inverse, see Rappaport and Silverman (1971). Eq. (3) is known as the generalised discrete-time algebraic Riccati equation . The with the additional constraint (4) is sometimes referred to as the constrained generalised discrete-time algebraic Riccati equation, herein denoted by . It is obvious that (3) is a generalisation of the classic , in the sense that any solution of is also a solution of –and therefore also of because –but the vice-versa is not true in general. Despite its generality, this type of Riccati equation has only been marginally studied in the monographs (Abou-Kandil et al., 2003, Ionescu et al., 1999, Saberi et al., 1995) and in the paper (Ferrante, 2004). The only contributions entirely devoted to the study of the solutions of this equation are Ionescu and Oaraˇ (1996) and Stoorvogel and Saberi (1998). The former investigates conditions under which the admits a stabilising solution in terms of the deflating subspaces of the extended symplectic pencil. The latter studies the connection between the solutions of this equation and the rank-minimising solutions of the so-called Riccati linear matrix inequality. In pursuing this task, the authors of Stoorvogel and Saberi (1998) derived a series of results that shed some light on the fundamental role played by the term . An example is the important observation according to which the inertia of this matrix is independent of the solution of , Stoorvogel and Saberi (1998, Theorem 2.4). Hence, (i) if is a solution of , then all solutions of will also satisfy and, (ii) if is a solution of such that is singular, then does not admit solutions. The results presented in Stoorvogel and Saberi (1998) are established in the very general setting in which the Popov matrix is not necessarily positive semidefinite as in (2).
It is often taken for granted that generalises the standard in the solution of the infinite LQ optimal control problem in the same way in which (Rappaport & Silverman, 1971) established that the generalised Riccati difference equation generalises the standard Riccati difference equation in the solution of the finite-horizon LQ problem. However, to the best of the authors’ knowledge, this fact has never been presented in a direct, self-contained and rigorous way. Thus, the first aim of this paper is to show the connection of the and the solution of the standard infinite-horizon LQ optimal control problem. The second aim of this paper is to provide a geometric picture describing the structure of the solutions of the in terms of the output nulling subspaces and the corresponding reachability subspaces of . Indeed, when , the null-space of is independent of the solution of , and is linked to the presence of a subspace which plays an important role in the characterisation of the solutions of and in the solution of the related optimal control problem. This subspace does not depend on the particular solution , nor does the closed-loop matrix restricted to this subspace. This new geometric analysis reveals that the spectrum of the closed-loop system is divided into a part that depends on the solution of , and one–coinciding with the eigenvalues of the closed-loop restricted to this subspace–which is independent of it. At first sight, this fact seems to constitute a limitation in the design of the optimal feedback, because regardless of the solution of the generalised Riccati equation chosen for the implementation of the optimal feedback, the closed-loop matrix will always present a certain fixed eigenstructure as part of its spectrum. However, when is singular, the set of optimal controls presents a further degree of freedom–which is also identified in Saberi et al. (1995, Remark 4.2.3)–that allows us to place all the closed-loop poles at the desired locations without changing the cost.
Section snippets
Linear quadratic optimal control and CGDARE
In this section we analyse the connections between LQ optimal control and CGDARE. Most of the results presented here are considered “common wisdom”. However, we have not been able to find a place where they have been explicitly derived, so we believe that this section may be useful. Consider the discrete linear time-invariant system governed by where and , and let the initial state be given. The problem is to find a sequence of inputs , with ,
Preliminary technical results
In this section, we present several technical results of independent interest that will be used in the sequel.
Geometric properties of the solutions of GDARE
Now we show that, given a solution of
- (a)
is an output-nulling subspace for the quadruple , i.e., ;
- (b)
the gain is such that is a friend of , i.e., .
Proposition 4.1
Let be the minimal positive semidefinite solution of . Then is the largest
Stabilisation
In the previous sections, we have observed that the eigenvalues of restricted to are independent of the solution of . This means that these eigenvalues are present in the closed-loop regardless of the solution of that we consider. On the other hand, coincides with the subspace , which is by definition the smallest -invariant subspace containing . Then, we can always find a matrix that assigns all the eigenvalues of restricted
Concluding remarks
In this paper we presented a self-contained analysis of some structural properties of the CGDARE that arises in infinite-horizon discrete LQ optimal control. The considerations that emerged from this analysis revealed that a subspace can be identified that is independent of the particular solution of CGDARE considered. Even more importantly, it has been shown that the closed-loop matrix restricted to this subspace does not depend on the particular solution of CGDARE, and has been shown to be
Augusto Ferrante was born in Piove di Sacco, Italy, on August 5, 1967. He received his “Laurea” degree, cum laude, in Electrical Engineering in 1991 and his Ph.D. in Control Systems Engineering in 1995, both from the University of Padova. He has been on the faculty of the Colleges of Engineering of the University of Udine, and of the “Politecnico di Milano” and visiting Professor at Curtin University in Perth (Australia). He is presently Professor in the “Department of Information Engineering”
References (14)
- et al.
Algebraic Riccati equation and -spectral factorization for estimation
Systems & Control Letters
(2004) - et al.
A unified approach to finite-horizon generalized LQ optimal control problems for discrete-time systems
Linear Algebra and its Applications
(2007) - et al.
The discrete-time algebraic Riccati equation and linear matrix inequality
Linear Algebra and its Applications
(1998) - et al.
Matrix Riccati equations in control and systems theory
(2003) On the structure of the solution of discrete-time algebraic Riccati equation with singular closed-loop matrix
IEEE Transactions on Automatic Control
(2004)- et al.
Order reduction of discrete-time algebraic Riccati equations with singular closed-loop matrix
Operators and Matrices
(2007) - et al.
Generalized discrete-time Riccati theory
SIAM Journal on Control and Optimization
(1996)
Cited by (0)
Augusto Ferrante was born in Piove di Sacco, Italy, on August 5, 1967. He received his “Laurea” degree, cum laude, in Electrical Engineering in 1991 and his Ph.D. in Control Systems Engineering in 1995, both from the University of Padova. He has been on the faculty of the Colleges of Engineering of the University of Udine, and of the “Politecnico di Milano” and visiting Professor at Curtin University in Perth (Australia). He is presently Professor in the “Department of Information Engineering” of the University of Padova. His research interests are in the areas of linear systems, spectral estimation, optimal control and optimal filtering, quantum control, and stochastic realisation.
Lorenzo Ntogramatzidis received his “Laurea” degree, cum laude, in Computer Engineering in 2001 from the University of Bologna, Italy. He received his Ph.D. in Control and Operations Research in 2005. From 2005 to 2008, he was a post-doctoral Research Fellow at the Department of Electrical and Electronic Engineering, The University of Melbourne, Australia. Since 2009, he has been with the Department of Mathematics and Statistics at Curtin University, Perth, Australia, where he is currently Senior Lecturer. His research interests are in the area of systems and control theory.
- ☆
Partially supported by the Italian Ministry for Education and Research (MIUR) under PRIN grant n. 20085FFJ2Z and by the Australian Research Council under the grant FT120100604. The material in this paper was not presented at any conference. This paper was recommended for publication in revised form by Associate Editor Faryar Jabbari under the direction of Editor Roberto Tempo.
- 1
Research carried out while visiting Curtin University, Perth (WA), Australia. Tel.: +39 049 8277681; fax: +39 049 8277699.