Hisaya Fujioka

A hybrid state-space approach to sampled-data feedback control

This paper develops a state-space theory for the study of linear shift-invariant finite-dimensional hybrid dynamical systems. By hybrid system, we mean an inputoutput operator relating hybrid signals, that is, signals which consist of two parts: a continuous-time part and a discrete-time part. We first introduce the hybrid state-space model and show the closedness properties under the fundamental operations. We next investigate the standard basic issues associated with this formalism: reachability, observability, and internal stability. We define the notion of Hz-norm of the hybrid system and show how to evaluate it. Since the analysis result can be applied to any hybrid system with the proposed hybrid state-space model, we apply the analysis result to the optimal digital control problem for standard four-block configuration with H,-norm performance measure. The solution, with computational algorithm, is derived for the time-delay case.

Modern and classical analysis/synthesis methods in sampled-data control-a brief overview with numerical examples

This paper gives a brief overview of modern analysis/synthesis methods for sampled-data feedback control and exhibits their advantages by numerical examples.

Computational aspects of gain-frequency response for sampled-data systems

This paper addresses the computational aspects of the frequency response for sampled-data systems. We first give two types of lower bound formulas. The first is derived via fast sampling of the input functions and the second is derived via the dual system. It is shown that as the fast sampling period approaches zero, these lower bounds converge to the gain. Two different formulas for upper bounds are also obtained. One is based on the information about the closed-loop poles, and another relates the gains of the /spl Hscr//sub 2/ norm equivalent discrete-time system. We then discuss the bisection search. It is seen that where certain non-degeneracy condition holds, bisection search is possible, some comparisons are made with numerical examples.

State covariance assignment problem with measurement noise: a unified approach based on a symmetric matrix equation

Abstract We consider a state covariance assignment problem with measurement noise. We derive a unified solution for both the continuous-time and the discrete-time case, based on a symmetric matrix equation of quadratic type. The solution includes (1) the existence condition for a feedback gain which assigns a specified covariance, (2) an explicit parametrization of the entire set of feedback gains that assign a specified covariance. We can show that the previous results for free-measurement noise can be readily derived as special cases.

Perspectives in Mathematical System Theory, Control, and Signal Processing

Circuit Theory.- Old and New Directions of Research in System Theory.- Regular Positive-Real Functions and the Classification of Transformerless Series-Parallel Networks.- Ports and Terminals.- Control.- On the Sample Complexity of Probabilistic Analysis and Design Methods.- A Constant Factor Approximation Algorithm for Event-Based Sampling.- A Unified Approach to Decentralized Cooperative Control for Large-Scale Networked Dynamical Systems.- Control and Stabilization of Linear Equation Solvers.- Nonlinear Output Regulation: A Unified Design Philosophy.- An Estimated General Cross Validation Function for Periodic Control Theoretic Smoothing Splines.- Stable Controller Design for Systems with Time Delays.- Dynamic Quantization for Control.- Graphs and Networks.- Convergence of Periodic Gossiping Algorithms.- Distributed PageRank Computation with Link Failures.- Predicting Synchrony in a Simple Neuronal Network.- Mathematical System Theory.- On the Stability and Instability of Pade Approximants.- On the Use of Functional Models in Model Reduction.- Quadratic Performance Verification for Boundary Value Systems.- Lyapunov Stability Analysis of Higher-Order 2-D Systems.- From Lifting to System Transformation.- Contractive Systems with Inputs.- Dissipativity and Stability Analysis Using Rational Quadratic Differential Forms.- On Behavioral Equivalence of Rational Representations.- Modeling.- Modeling and Stability Analysis of Controlled Passive Walking.- An Optimization Approach to Weak Approximation of Levy-Driven Stochastic Differential Equations.- Compound Control: Capturing Multivariable Nature of Biological Control.- Law of Large Numbers, Heavy-Tailed Distributions, and the Recent Financial Crisis.- Signal Processing.- Markov Models for Coherent Signals: Extrapolation in the Frequency Domain.- Digital Signal Processing and the YY Filter.- Sparse Blind Source Separation via ?1-Norm Optimization.- YY Filter - A Paradigm of Digital Signal Processing.- System Identification.- How to Sample Linear Mechanical Systems.- A Note on LQ Decomposition in Stochastic Subspace Identification.- Modeling Systems Based on Noisy Frequency and Time Domain Measurements.- Blind Identification of Polynomial Matrix Fraction with Applications.

State covariance assignment for sampled-data feedback control systems

This paper proposes a new performance criterion of ‘covariance’ for sampled-data systems. A covariance of sampled-data systems is defined by taking account of inter-sample behaviour. An SCA (state covariance assignment) problem for sampled-data feedback control systems is also discussed, which is the counterpart of that for purely continuous or discrete-time feedback control systems. The SCA problem for sampled-data systems will be solved as a discrete-time SCA problem, where the discretization preserves the state covariance and is in two steps. In the first step, a certain sample time performance is required to ensure the inter-sample performance, and the output signal results to be discretized. The second step is to discretize the input signal.

Robust performance synthesis for sampled-data feedback control systems

The constantly scaled sampled-data /spl Hscr//sub /spl infin// control problem is discussed for robust performance analysis and synthesis. Based on the linear operator inequality characterization of the sampled-data /spl Hscr//sub /spl infin// norm bounding condition, the convex properties in analysis and state feedback synthesis problems are shown. A finite step global optimization algorithm is proposed for output feedback synthesis problem.

Norm bound preserving disturbance input subspaces in sampled-data /spl Hscr//sub /spl infin// control problem

The NBP (norm bound preserving) subspace in the sampled-data /spl Hscr//sub /spl infin// control problem is investigated The mutual interrelation among the existing solutions for the problem is given as an inclusion relation among the corresponding NBP subspaces. Two new equivalent systems are also derived by extending the NBP subspaces related to the existing solutions.<<ETX>>