In this paper, we propose a new controller design method for decentralized systems. The method includes an arrangement scheme of multi-controllers and a controller design procedure via the LMI method. The arrangement scheme is based on analysis of a concept that we named ‘virtual hierarchical order’. In the scheme, we characterize the stable equilibrium value of a probability matrix as the dense of the connection and determine the order of the subsystem. Multi-controllers are arranged according to the virtual hierarchical order and constitute a decentralized control system.
The controller design utilizes the function of LMI based methods, which enables one to put constraints on the form of the feedback controller. By applying a change of variables method and setting up the structure of variables to be the same as the controller structure, a desired controller is obtained. We show a design procedure of a structured controller. The validity of proposed methods is examined on a numerical model.

View full abstract

Differential equations for the weighting matrix on the states have been derived from the characteristic equation of the Hamiltonian matrix of the linear quadratic regulator (LQR). The differential equations are integrated from an initial set of closed-loop eigenvalues to an assigned one, which yields a weighting matrix for the desired eigenvalues. In the case where an indefinite weighting matrix results, another set of differential equations derived from the characteristic equation of the weighting matrix are integrated from negative eigenvalues to zero or positive values along with the differential equations for the closed-loop eigenvalues. This manipulation makes the indefinite matrix nonnegative definite. Furthermore, differential equations are derived from the equation that defines eigenvectors for corresponding eigenvalues. Integrating the differential equations along with the above ones provides a nonnegative definite weighting matrix that achieves eigenstructure assignment. However, it should be noted that this is not always possible, depending on assigned eigenvalues and/or eigenvectors. The effectiveness of the proposed method is illustrated through applications to a simple example and to linearized equations of motion of the F-4 aircraft.

View full abstract

This paper presents a stabilizing controller design br an affine input system with discontinuity. We pursue the uniqueness of the Filippov solution of the closed loop system in addition to the asymptotic stability. First, a typical L_gV structure-based switching controller is considered. However, it is shown that the uniqueness of the solution of the closed loop system is not necessarily assured by the controller in the whole state space. Then, an asymptotic stabilizing controller which entirely assures the uniqueness in the whole state space is developed by modifying the L_gV structure-based switching controller.

View full abstract

This paper studies a higher order extension of Stokes-Dirac structures. The Stokes-Dirac structure is defined by differential forms on a spatial domain of systems and its boundary. The structure has a power-conserving property, that is the change of the interior energy is equal to the power supplied to the system through the boundary. The Stokes-Dirac structure is used for port-Hamiltonian formulations of distributed-parameter systems mainly. The higher order representation does not only use exterior derivative operators, but also both Hodge star operators and their composite operators to relate flows to efforts. The extended structure expresses differential relations between energy variables, and global properties of systems through harmonic forms. First, some mathematical preparations are introduced to define the higher order energy variables and the arbitrariness of the representation is explained. Next, three types of the higher order Stokes-Dirac structures are proposed. Finally, two examples: Euler-Bernoulli beam models and Electromagnetic wave equations are presented.

View full abstract

This paper deals with a design method of compensation law for linear constrained systems. By employing singular value decomposition of linear systems, we provide a design method of compensation law which fulfills system constraints. The feature of resulting compensation law is illustrated with numerical examples.

View full abstract

Self-organizing map-based methods for the detection of confusion between blood test data are presented. Learning data for the self-organizing map (SOM) is generated by subtracting each element of complete blood count (CBC) data of the immediately previous patient's results from that of the current results. The neurons in the well-trained SOM are roughly divided into two clusters: one with neurons reacting to regular input data, and the other with neurons reacting to irregular input data generated by subtraction between confused CBC data. If a winner neuron belongs to the latter cluster, it is presumed that confusion has arisen between the CBC data of different patients. In addition, a genetic algorithm is adopted to eliminate redundant elements in the CBC data, which have an unfavorable influence on the judgment of confusion. Experimental results show that the proposed methods achieve high accuracy of detection even when the input data irrelevant to the learning of maps is applied to them.

View full abstract

To investigate the performance of sound source localization in the front-middle plane at arbitrary body positional angles, a mobile three-D acoustic stimulator was developed. The angle of the body position is adjustable in parallel with the sound sources that are arranged around the subjects. In the present study, 4 identical speakers were arranged in the front-middle plane of the subject. The speakers were separated by 9 degrees in the vertical plane. Using a white noise tone (0.3-s duration), a sound source discrimination task was conducted in five body positions at 90-, 67.5-, 45-, 22.5- and 0-degree angles. As a result, the average of the correct response rate was higher in the 90- and 0-degree body positions than that in the 67.5-, 45- and 22.5-degree body positions. The localization of the source position tended to be distorted (skewed) to the middle (eye) level in all body positions. The vertical distortion to the middle (eye) level became large in the 67.5-, 45- and 22.5-degree body positions compared to that in the 90- and 0-degree body positions. These results suggest that sound source localization in the front-middle plane is affected not only by the body position but also by the vertical sound position.

View full abstract

A recursive maximum filter (RMF) is an algorithm devised to solve the problem of detecting small moving targets in noisy image sequences. RMF is simple and effective for the enhancement of small moving targets with a low signal-to-noise ratio; however, its principle and performance limit are not clear because the algorithm was derived heuristically. In this paper, we reformulate RMF based on a Bayes estimation and show that it can be interpreted as a Bellman equation of dynamic programming (DP). Although some DP-based algorithms have already been proposed, RMF requires much less computation than previous algorithms because its state space is much smaller. RMF includes two design parameters: neighborhood size and a forgetting factor. We derive approximation formulae of the distributions of RMF outputs for various parameter values. By using the formulae, we show a minimum SNR with which targets are detectable for each neighborhood size. We also show the conditions under which targets can be detected by RMF with various parameter values.

View full abstract

We have proposed a system called Attentive Workbench (AWB) that is designed to support desktop work. AWB is the system which estimates a worker's condition using some sensors and assists a worker using movable trays and projectors from both physical and informational viewpoints. A method for estimating an AWB user's intention is proposed and presented herein. An AWB user expresses his/her intention by pointing a finger at a necessary tray. For estimating which tray a user needs, we take two approaches. One is to consider user's subjective pointing direction, and the other is to integrate the sensor data and user's past action sequences using Dynamic Bayesian Network. A description of the experimental results will demonstrate the usefulness of the method proposed here.

View full abstract

A method to construct a decentralized control law for large scale interconnected systems is proposed. The feedback gain is calculated with a solution of linear matrix inequalities. The resulting closed loop system belongs to a class of linear quadratic regulators.

View full abstract

This paper is an engineering study for the medication control in the medical field. It describes a medication control of steroid based on the data in 880 days using an LS model for a second order transfer function, a multiple correlation analysis and a crisis strategy control in the instrument and control engineering for ophthalmologists. The aim of this study is the identification of the control object and the identification of the dead time of the medication for each step of the medication using system identification for a second order transfer function and a multiple correlation analysis based on the data. The main point is how many data is suitable for above-mentioned aim. The input (actuating signal) of the model is the eye sights (visus RVCc and LVCc) and the output (controlled variable) is pieces of predonine 5mg par day. The eye sight as an measure of Opacitus Corporis virtrei (OCV). The first author has the blurred vision and the secondary contract and the secondary graucoma caused by the steroid syndrome. The paper also describes a crisis strategy control for the steroid syndrome.

View full abstract