H∞ control of discrete-time uncertain linear systems with quantized feedback☆
Introduction
In recent years, quantization in feedback control systems has attracted a growing interest [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]. This is mainly due to the wide application of digital computers in control systems and the rapid development of network based control. Comparing with classical control theory, quantized feedback control is a common source of errors, which may degrade the system performance as described by Kalman [1], the effect of quantization in a sampled data control system and pointed out that if a stabilizing controller was quantized using a finite-alphabet quantizer, the feedback system would exhibit limit cycles and chaotic behavior. Consequently, a lot of works have focused on understanding and mitigating the quantization effects in the early. While in recent studies, a general practice is to treat the quantizers as information coders. Among these results, there are mainly two approaches for studying control problem with quantized feedback. The first approach handles static quantizers such as uniform and logarithmic quantizers [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], while the second approach considers the dynamic quantizers which scales the quantization levels dynamically in order to improve the steady-state performance [13], [14].
For the problem of quantized feedback control, many important achievements have been obtained. Elia [2] had proven that a logarithmic quantizer is needed for stabilization of discrete-time single-input–single-output (SISO) linear time-invariant systems. Fu [3] gave a comprehensive study on feedback control systems with logarithmic quantizers by the sector bound approach. Both stabilization and performance issues have been considered. Following this work, Gao [7] noticed that the constant Lyapunov function is conservative for quantized feedback problem and proposed a new general framework based on quantization dependent Lyapunov functions. Recently, Zhou [8] revisited the absolute stability approach, and gave a less conservative result. For some new results about quantized feedback control, see [15], [16], [17]. Considerable attention, however, have been paid toward the study of control for linear systems [[18], [19], [20], [21], [22], [23], [24]], nonlinear systems [[25], [26], [27], [28]]. For polytopic uncertain systems, refs. [19], [20], [21], [22], [23] focus on the improved problem of the bounded real lemma (BRL) for the polytopic uncertainty systems, that is, how to find a less conservative LMI-based method of designing controller. It was noted that the basic idea behind these papers is based on constant feedback matrix K. However, the constant feedback matrix is independent of polytopic uncertainty parameters, the results obtained with constant feedback matrix are conservative when extended to polytopic uncertainty system with quantized feedback. Our main objective is to propose a new parameter dependent control law and to obtain less conservative results for polytopic uncertainty quantized feedback systems.
In this paper, we present a new way to deal with the quantized feedback problem for polytopic uncertainty systems, that is, change constant feedback matrix K into parameter dependent by Lagrange׳s interpolation. Obviously the result obtained by parameter dependent feedback matrix is less conservative than the ones by constant feedback matrix for polytopic uncertainty systems with quantized feedback. Finally, we will illustrate the effectiveness and reduced conservatism of our main results by a numerical example.
Notations: The symbol ⁎ induces a symmetric structure in LMIs. Generally, for a square matrix , denote its transpose and denotes . Matrices are assumed to have compatible dimensions.
Section snippets
Problem statement and preliminaries
Consider the following linear discrete-time system with polytopic uncertainties:where is the state variable, is the control input, is the control output and is the noise signal that is assumed to be the arbitrary signal in . The uncertain matrices , , , , and belong to the polyhedron
Main results
As we know well that for state feedback problem the constant feedback matrix K renders the condition to be conservative when matched with the polytopic uncertainties described in (2). Our main objective is to change K into parameter dependent by Lagrange׳s interpolation, i.e., by using Lagrange׳s interpolation estimate, the system parameter θ described in (2) further gives a new control law parameter dependent on the estimation of θ. First, with the quantized error considered, a robust
Numerical example
In this section, a numerical example is presented to illustrate the effectiveness of the proposed method. First, we will give an example for the parameter dependent control law design, then we will compare the proposed method with those appearing recently in the literature show that the proposed conditions provide less conservative results.
Assume that there data points are known, they are ; ; . Now, we can get the estimation of θ from the known data points by using Lagrange׳s
Conclusion
In this paper, the problem of robust control for the uncertain discrete-time system with quantized state feedback has been addressed. Based on parameter dependent Lyapunov function approach and Lagrange׳s interpolation, when we change the controller gains into parameter dependent, a less conservative condition has been proposed in terms of LMIs. Simulation comparative analysis has demonstrated the effectiveness and less conservative of the main results.
Zhi-Min Li received the B.E. degree form Beijing Union University, Beijing, China, in 2013. He is currently working toward the M.S. degree in the College of Engineering, Bohai University, Jinzhou, China. His research interests include robust control, networked control system, and fuzzy control.
References (31)
- et al.
A new approach to quantized feedback control systems
Automatica
(2008) - et al.
On the absolute stability approach to quantized feedback control
Automatica
(2010) - et al.
Attainability of the minimum data rate for stabilization of linear systems via logarithmic quantization
Automatica
(2011) - et al.
A sector bound approach to feedback control of nonlinear systems with state quantization
Automatica
(2012) - et al.
Robust quantized feedback stabilization of linear systems
Automatica
(2008) Hybrid feedback stabilization of systems with quantized signals
Automatica
(2003)- et al.
Stabilization of Markov jump linear systems using quantized state feedback
Automatica
(2010) - et al.
Consensus with quantized relative state measurements
Automatica
(2013) - et al.
Relaxed stability criteria for discrete-time Takagi–Sugeno fuzzy systems via new augmented nonquadratic Lyapunov functions
Neurocomputing
(2015) - et al.
Mixed and passivity based state estimation for fuzzy neural networks with Markovian-type estimator gain change
Neurocomputing
(2014)
control design for discrete-time switched fuzzy systems
Neurocomputing
Control synthesis of Roesser type discrete-time 2-D T–S fuzzy systems via a multi-instant fuzzy state-feedback control scheme
Neurocomputing
A stabilization algorithm for a class of uncertain linear systems
Syst. Control Lett.
Improved robust H2 and filtering for uncertain discrete-time systems
Automatica
Cited by (7)
Event-triggered H<inf>∞</inf> state estimation for semi-Markov jumping discrete-time neural networks with quantization
2018, Neural NetworksCitation Excerpt :Due to the storage limitation and communication bandwidth among nodes, the original precise information needs to be quantized (Zhang, Wang, Ding, & Liu, 2015). There has been a series of quantitative methods proposed in the previous literature (Hu & Yue, 2012; Li, Chang, Du, & Yu, 2016). The measurement output of the neural network is transmitted through an unreliable communication channel with certain constraints to the estimator.
Robust Passivity Analysis of Stochastic Genetic Regulatory Networks with Levy Noise
2022, International Journal of Control, Automation and SystemsH<inf>∞</inf> state estimation of discrete-time markov jump neural networks with general transition probabilities and output quantization
2017, Journal of Difference Equations and Applications
Zhi-Min Li received the B.E. degree form Beijing Union University, Beijing, China, in 2013. He is currently working toward the M.S. degree in the College of Engineering, Bohai University, Jinzhou, China. His research interests include robust control, networked control system, and fuzzy control.
Xiao-Heng Chang received the B.E. and M.S. degrees from Liaoning Technical University, Fuxin, China, in 1998 and 2004, respectively, and the Ph.D. degree from Northeastern University, Shenyang, China, in 2007. He is currently a Professor with the College of Engineering, Bohai University, Jinzhou, China. His research interests include fuzzy systems and uncertain systems.
Xiao-Kun Du received the B.E. and M.S. degrees from Liaoning Technical University, Fuxin, China, in 2003 and 2006, respectively. She is currently pursuing the Ph.D. degree at the School of Electrical Engineering and Automation, Tianjin University, Tianjin. Her research interests include fuzzy systems and uncertain systems.
Lu Yu received the B.E. degree form Bohai University, Jinzhou, China, in 2008. He is currently working toward the M.S. degree in the Software College, Northeastern University, Shenyang, China. His research interests include networked control system and fuzzy control.
- ☆
This work was supported in part by the National Nature Science Foundation of China under Grant 61104071, the Program for Liaoning Excellent Talents in University, China under Grant LJQ2012095, the Open Program of the Key Laboratory of Manufacturing Industrial Integrated Automation, Shenyang University, China under Grant 1120211415.