In attaining a high-aged and high-welfare society, various demands have been raised on health monitoring. One demand is to develop an unconstrained and noninvasive measurement instrument for respiration and heart rates. This is because they are the most basic information in monitoring the health of high-aged persons or patients in daily lives. Along this objective, several methods have been studied. But, the methods not only aimed to measure only one signal of the respiration and heart rates, but also were not the methods to assure a complete unconstrained measurement. Moreover, the methods needed much cost in making the equipment.
The paper thus proposes a convenient measurement system which makes use of a wire-type respiration pick-up pasted on an air-mat (or water-mat), on which the subjects are laid down, and extracts the respiration and heart rates by applying low-pass and high-pass filters, respectively, to the data. Experiments show that the method offers accurate measurement of respiration and heart rates except for the duration of body movement. The measurement system is thus not only effective, but also easy to use for the cost, the size and the usage of the unconstrained and noninvasive type.

抄録全体を表示

The applications using the optical measurement systems often require the measurement performance to be fast, of high accuracy, and capable of simultaneously detecting plural points. In this paper, a new method has been developed to measure the positions of plural light spots in a continuous manner using a sensor array, which consists of plural sensor elements. The basic principle of the method detects the sensor elements that generate the peak values by comparing the outputs of the sensor elements with the standard voltage, and then measures the actual peak positions with the use of an analog interpolation. This method is different from others in that it uses an analog processor for peak detection and calculation algorithm. The method allows the simultaneous processing of the peak detection and the calculation of the interpolation. As a result, this method can meet three requirements that are often required of this kind of measuring system: (1) high-speed measurement, (2) high-accuracy performance, and (3) simultaneous measurement of plural points. We have designed and built a prototype position sensor system that implements the proposed method. The experimental results are as follows: (1) the resolution of 1μm or less; (2) the sampling time is 6μs/point; (3) simultaneous measurement of 4 points.

抄録全体を表示

This paper presents a novel approach to the robust H_∞ control problem for uncertian nonlinear systems with stable zero-dynamics. The uncertainty is described by unknown but gain bounded functions of the state variables. A parametrization of a class of robust stabilizing control laws is derived for the uncertain systems, and a solution to the robust H_∞ control problem is obtained by constructing a storage function based on the Lyapunov function of robustly stable zero-dynamics.

抄録全体を表示

In this paper, we propose a numerical solution to a class of parameter-dependent convex differential inequalities (PDCDIs), including parameter-dependent LMIs formulated in analysis and synthesis of control systems based on linear parameter varying system representations. To solve such inequality problems involving infinite inequalities corresponding to the values of the parameter, we provide a procedure to derive from a given PDCDI a finite set of inequalities that can be formed to be solvable whenever the PDCDI is solvable, and from any solution to the derived finite set of inequalities a solution to the PDCDI is obtained. Numerical examples for a parameter-dependent LMI are provided to examine the proposed numerical solution.

抄録全体を表示

In this paper the Hamilton-Jacobi equation in nonlinear control theory is analyzed using symplectic geometry. First, a solution method of the Hamilton-Jacobi equation is outlined and shown to be a natural extension of the well-known theory of the Riccati equation. Second, a necessary and sufficient condition for the existence of the stabilizing solution is proposed. Finally, the structure of solutions of the Hamilton-Jacobi equation, such as the maximal and minimal solution, is clarified. To this end, the theory of the nonlinear Lyapunov equation is developed.

抄録全体を表示

In this paper, we discuss structural decoupling conditions for generalized multivariable systems. Some designing proceduree for decouple control have been investigated by many authors, but most decoupable condition depend on system parameters. It was shown that the structural decoupling condition could be derived with the aid of graph-representation by Linnemann for state variable systems.
The descriptor variable representation has an advantage that it can reflect physical structure of real systems. Therefore, a new decoupling condition for descriptor systems is derived via modified bipartite graphs. Further, it is shown that propsed results include the Linnemann's result.

抄録全体を表示

This paper considers the problem of impulse response identification for a linear sampled-data system where the input signal is held constantly within a multiple of the sampling period of the output signal. It is well known that in this case, the identification problem is usually ill-conditioned, if the system under study is discretized with a much shorter sampling period of the output signal. To improve the identification accuracy, this paper proposes a new identification approach by employing the Haar scaling and wavelet functions. At first, we point out that the discrete-time impulse response of a linear system with zero-order hold (ZOH) input is actually the piecewise-constant approximation of its continuous-time impulse response. Then based on the close relation between piecewise-constant approximation with Haar scaling and wavelet functions, a hierarchical identification procedure is proposed which identifies the system impulse response from a coarse resolution level (low frequency subspace) to a fine resolution level (high frequency subspace) successively. At each resolution level, the BIC is utilized to determine the length of the decomposed impulse response in the corresponding subspace, so that some redundant parameters in the high frequency-domain which are sensitive to the noise effects are discarded. Since the identified impulse response model is not smooth when it is represented by some Haar scaling functions of different widths, we can replace each Haar scaling function in the preidentified impulse response model by a gaussian basis function with corresponding position and width. Then an improved identification method is also proposed to achieve smooth continuous-time impulse response model from sampled data, based on the designed gaussian basis functions. It is shown through simulation study that the proposed method yields accurate estimate of the impulse response even in the ill-conditioned cases.

抄録全体を表示

The dynamics solution for contact state transition of three-dimensional multi-rigid body manipulation systems with Coulomb friction is proposed. A linear complementarity problem method is applied to obtain governing equations for the dynamics model of the multi-rigid body manipulation systems. The conditions under which the multi-rigid body manipulation systems are static determinancy are derived. Algorithms for solving these problems using Lemke's algorithm are shown. With an illustrating numerical example, the effectiveness of the computational scheme is verified and the contact state transitions for friction coefficients and joint torques are discussed.

抄録全体を表示

Manufacturing systems become complex more and more for adapting to various environmental conditions. In these systems, sensor fusion methods for estimating states of a system from multiple sensor information have received much attention. Also, it is necessary to select the sensor infromation for adapting to various situation flexibly. In these days, various methods of fusing multiple information have been proposed so far, but these methods don't select the sensor information by the reliability of each sensor. In this paper, we propose a sensor selected fusion system using a recurrent neural network. The sensor selection method is based on the production system considering with the reliability which expressed by the possibility measure. The effectiveness of the proposed method is shown through inferring the surface roughness in the grinding process.

抄録全体を表示

In this paper, we propose an evolutionary approach to architecture design of modular dynamical neural networks. As one of modular dynamical neural networks, we adopt Cross-Coupled Hopfield Nets (CCHN) in which plural Hopfield networks are coupled to each other. The architecture of CCHN is represented by some structural-parameters such as the number of modules, the numbers of units per module, the module connectivity, and so forth. In the proposed design method, these structural-parameters are treated as phenotype of an individual, and suitable modular architecture is searched through the evolution of its genetic representation (genotype) by using genetic algorithms. Based on a simple direct coding method, the order of length of genetic representation for the structural-parameters can be estimated to be O(N²) where N is the total number of units. On the other hand, the order of genetic representation proposed here is O(N). To verify the usefulness of proposed method, we apply a CCHN to associative memories. Here, the fitness of an individual is defined so as to be larger when a CCHN has a simpler architecture as well as when the association performance is higher. As the result of simulations, we certify that the proposed design method can find high-performance CCHN with simple modular architectures.

抄録全体を表示

The purpose of this study was to investigate the transfer between vision and proprioception. In our previous study, we designed two tasks, a VV task and a VP task (V: visual, P: proprioceptive), in wich the input is given by the visual information and output is measured by the visual or proprioceptive value. The results showed that the transfer from visual to proprioceptive inner space can be approximated as an affine transfer. Information concerning location was imprecise and the shape expanded by 20 percent, but linearity was maintained. In this study, we designed two tasks: a PP task and a PV task. In both tasks, two proprioceptive targets were indexed by the examiner to the subject by using finger of the index limb. In the PP task, we investigated the internal transfer of proprioception. The subject indicated the midpoint of two targets with the finger of the indicating limb that is contralateral of index limb using only proprioceptive information. In the PV task, we investigated the internal transfer from proprioception to vision. The subject indicated the midpoint using a visual pointer. The results showed that linearity was not maintained in either the PP or PV task. The transformation from proprioception could not be apporximated as an affine transformation on the Cartesian coordinates. When we compared our data under three coordinate models (Cartesian, polar, and kinesthetic coordinate models), we found that the kinesthetic coordinate model enabled the greatest reproducibility; i.e., the kinesthetic coordinate model showed the smallest coefficient of variation (Cv). In the case of describing to the proprioceptive information, the kinesthetic coordinate system enabled the greatest degree of data reproducibility.

抄録全体を表示