Lyapunov-like Stability Inequality with an Asymmetric Matrix and application to Suboptimal Linear Quadratic Control
Avinash Kumar [email protected] · 2025
Read Lyapunov-like Stability Inequality with an Asymmetric Matrix and application to Suboptimal Linear Quadratic Control by Avinash Kumar [email protected] (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-11556.txt.
Introduction
Avinash Kumar presents a 2025 paper on a new form of the Lyapunov inequality, a core tool for proving that a control system is stable. The classic inequality demands a symmetric positive matrix, which certifies that a system settles rather than blows up. Kumar asks what happens if that matrix may be asymmetric. He studies the variant's properties for a class of systems and derives fresh stabilization results.
He then applies the theory to suboptimal linear quadratic control, a design problem balancing performance against control effort. The asymmetric matrix doubles as the design tool, and he characterizes the cost of the resulting control laws with an upper bound tied to starting conditions. Two examples show the method at work: one control design and one agreement protocol for a team of agents. In the second, asymmetry turns out to be required, not optional.
Catalog record
| CREATOR | Avinash Kumar [email protected] |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-11556
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Lyapunov-like Stability Inequality with an Asymmetric Matrix and application to Suboptimal Linear Quadratic Control Avinash Kumar [email protected] Abstract The Lyapunov inequality is an indispensable tool for stability analysis in linear control theory. It provides a necessary and sufficient condition for the stability of an autonomous linear-time invariant system in terms of the existence of a symmetric positive-definite Lyapunov matrix. This work proposes a new variant of this inequality in which the constituent Lyapunov matrix is allowed to be asymmetric. After analysing the properties of the proposed inequality for a class of matrices, we derive new results for the stabilisation of linear systems. Subsequently, we utilize the developed results to obtain sufficient conditions for the suboptimal linear quadratic control design problem wherein addition to having an asymmetric Lyapunov matrix, which serves as a design matrix for this problem, we provide a characterization of the cost associated with the computed stabilizing suboptimal control laws. This characterization is done by deriving an expression for the upper bound on cost in terms of the initial conditions of the system. We demonstrate the applicability of the proposed results using two numerical examples- one for suboptimal control design for a linear time-invariant system and another for the consensus (state-agreement) protocol design for a multi-agent system, wherein we see how the asymmetry of the Lyapunov (design) matrix emerges as an inherent requirement for the problem. 1 Introduction Lyapunov theory is a vital tool for stability analysis in control theory, and its importance cannot be overstated. Lyapunov inequality, which results upon application of this theory to the linear time-invariant (LTI) systems with a quadratic positive-definite Lyapunov function, serves as the backbone for studying the stability properties of linear systems. It finds applications not only in the linear control theory but also in the robust and nonlinear control design [ 1 ] , linear systems with disturbances [ 2 ] , and model predictive control [ 3 ] . It is the first linear matrix inequality (LMI) to appear in the control theory [ 4 ] ; the field of LMIs has been advancing continually since then [ 5 ] . The LMIs are a vital tool for the control design techniques today, thanks to easy solvability owing to the availability of computational resources [ 6 ] . As per the author’s best knowledge, while utilizing Lyapunov inequality, the symmetry of the constituent Lyapunov matrix is always assumed. See the box labeled “ The Symmetry Assumption ” for the reasons for this pervasive assumption. This work presents new results for a class of matrices wherein this assumption is dropped. The reasons for the presence of the symmetry assumption are vivid [ 7 ] - 1. the nice properties that symmetric matrices offer, namely, possession of real eigenvalues and the property of diagonalisability; 2. the quadratic function, say \mathbf{z}^{T}M\mathbf{z} with \mathbf{z}\in\mathbb{R}^{n} and a symmetric M\in\mathbb{R}^{n\times n} , satisfies \mathbf{z}^{T}M\mathbf{z}=\mathbf{z}^{T}M_{sym}\mathbf{z} where M_{sym}=\frac{1}{2}\left(M+M^{T}\right) is the symmetric part of the matrix M ; and 3. the notion of the positive/negative-definiteness of the scalar quadratic from \mathbf{z}^{T}M\mathbf{z} is easily transformed to an equivalent notion of positive/negative-definiteness of the constituent symmetric matrix M . The Symmetry Assumption Linear-quadratic (LQ) optimal control design is a well-developed theory, wherein the optimal solution is obtained by solving the Riccati equation [ 8 , 9 ] . Nevertheless, the optimal solution may not always be implementable, because of the saturation constraints on the input, norm-boundedness requirements on the feedback-gain matrix, because this norm directly augments the maximum control effort [ 10 ] , or the structural conditions on the feedback-gain matrix [ 11 ] . Thus, it becomes important to compute and analyse the suboptimal solutions for this problem. In the existing literature, to approach this suboptimal control design problem, an upper bound is pre-specified, and then sufficient conditions for the existence of suitable suboptimal control laws are derived [ 12 ] . These conditions impose that the initial conditions of the system reside in a certain set (ball); contrariwise, in this work, we develop results utilising the exact value of initial conditions and then argue that this latter technique can be extended straightforwardly to the former approach. Moreover, specifying the upper bound a priori may lead to infeasibilities if the upper bound is not chosen cautiously, especially for the networked control design problems, where the optimal cost is not known. In this work, we develop new results based on the proposed inequality and then utilize them to compute the stabilizing control law as well as the associated upper bound on the cost in terms of the initial conditions of the system [ 11 ] . Finally, using suitable numerical examples, we demonstrate the usage of these results to suboptimal LQ control design and to suboptimal consensus protocol design for a multi-agent system. We demonstrate in the latter example that the problem necessitates the computation of an asymmetric Lyapunov matrix, which serves as a design matrix for this problem. In summary, the contributions of this work are as follows. 1. It puts forward a new Lyapunov-like stability inequality for a class of matrices and the properties thereof; 2. it develops new results for the characterisation of the cost for suboptimal LQ control laws, where the upper-bound on the cost is computed in terms of the initial conditions, instead of specifying it a priori as generally done in the literature. The results proposed in this paper are expected to open up a new line of research in LMIs where the constituent Lyapunov matrices are not necessarily assumed to be symmetric thus allowing the enlargement of the search space for these design matrices. The remainder of the paper is organized as follows. In Section 2 , the background of the Lyapunov inequality, followed by the suboptimal linear quadratic control design problem, is presented; the remarks in this section collect the important observations regarding the results present in the literature, which which partly serve as a basis for the study carried out. Section 3 presents the main results of the paper -namely, a new Lyapunov-like inequality, a result for stabilisation of an LTI system, followed by the quantification of the suboptimal LQ control laws. Section 4 presents the numerical results, followed by the concluding remarks in Section 5 . N otation: Throughout this manuscript, scalars and scalar-valued functions are denoted by small ordinary alphabets, vectors are denoted by small bold alphabets, and capital ordinary alphabets denote matrices. M\succ\mathbf{0}(\succeq\mathbf{0}) and M\prec\mathbf{0}(\preceq\mathbf{0}) means that the matrix M\in\mathbb{R}^{n\times n} is symmetric positive- (semi-)definite and symmetric negative-(semi-)definite respectively. ‘ ⋆ ’ in the superscript of a quantity denotes the conjugation, ‘ T ’ and ‘ H ’ in the superscript denote transposition and the conjugate transposition respectively. I_{n\times n} denotes an identity matrix of order n . A ‘ ∗ ’ in superscript denotes the optimal counterpart of the quantity under consideration. trace(M) , \lambda(M) and \lambda_{i}(M) denote the trace, an eigenvalue and i^{th} eigenvalue of the matrix M respectively. Re(m) denotes the real part of the quantity m . 2 Background and Problem Formulation In this section, we present the background of Lyapunov inequality, existing approaches to the suboptimal LQ control design problem, and the insights into the results that we shall develop, followed by