The purpose of this study is to investigate effects of vigilance states on the characteristics of visually triggered saccade quantitatively. In this paper saccades and EEG at three position on the scalp of Fz, Cz and Pz have been measured in two different experimental conditions: target discrimination task (Task) and Non-Task. In Task condition subjects chased a target, then discriminated the figure of it and put the button when the target was triangle. In Non-Task condition subjects only chased a spot target. EEG data obtained from both conditions were analyzed by means of Fast Fourier Transform and two parameters of vigilance states were extracted from the spectral power of EEG. Comparison of the vigilance state parameters and the saccade parameters between two conditions showed that in Task condition, vigilance states were higher, saccade latencies were shorter, saccade absolute position errors were smaller, saccade durations were shorter, and saccade peak velocities were faster than in Non-Task condition. These results imply that the rise in vigilance state in Task condition may have changed the characteristics of saccade. In order to examine this possibility, data of saccades were divided to high vigilance group and low vigilance group according to the value of EEG parameters. The results indicated that saccade latencies and durations became shorter and saccade peak velocities faster due to the rise in vigilance state. These facts suggest that new EEG parameters we proposed reflect vigilance states quantitatively and these are practical for analysis of saccades, and that the characteristics of saccade have a dependence on vigilance states.

抄録全体を表示

The triple point of diphenyl ether is realized using small pyrex glass cells (35mm in diameter and 120mm in length). Commercially available diphenyl ether with the nominal purity of 99.0% is purified by the method of distillation and the zone-melting technique up to 99.998%, which is estimated from the melting curves by the thermal analysis. The solid-liquid phase transition temperature is determined to be 26.8633 (0.001) based on the International Temperature Scale of 1990. It is also found that the width of the melting curve and the length of the plateau depend on the freezing temperatures.

抄録全体を表示

In this paper, we consider the design problem of simultaneous stabilizing controller which takes into account various closed-loop performances, e.g. sensitivity or robust stability etc.. Using the parametrization of simultaneous stabilizing controller, this problem is formulated as a standard model-matching-type problem whose objective function is affine in the design parameter. Under this formulation, a simple sufficient condition to achieve simultaneous stabilization is derived as in the form of an inequality H_∞ norm constraint on the design parameter. Moreover, we show that the presented design problem including above constraint can be cast as convex optimization problem. By a design example, the effectiveness of the proposed method is presented.

抄録全体を表示

This paper proposes a generic design method to develop an optimal state feed-back control system using multilayered neural networks. Because Back-Propagation can not be applied to this case, two new methods are proposed for training neural networks: one is based on gradient method and the other on Powell's conjugate direction method. These two methods can be applied to various problems such as ones for non-linear systems where the feed-back controller can not be obtained by conventional methods. Especially, the method based on the direction method can deal with criterion functions whose derivatives can not be obtained analytically. Illustrative examples of optimal regulator and minimal time control problems show the effectiveness of the proposed methods.

抄録全体を表示

In this paper, for the two dimensional linear control system described by the transfer function, a design method of the unity feedback control system using the fuzzy partition of the design plane is proposed. That is, according to the design specifications, we fuzzily part the design plane and assign the gain which gives characteristic roots satisfying the design specifications on each partition. And, we infere the whole gain of the system from the each gain by the fuzzy reasoning. Then, we obtain the simulation results of the systems with the special damping ratio ζ of the characteristic roots. One of the damping ratio is chosen from Butterworth pattern, the other is ζ=(1.5, 0.2). Next, using those simulation results, we show a design method of the unity feedback control system. Lastly, we apply this design method to the fuzzy compensation element design.

抄録全体を表示

This paper deals with the nD practical tracking problem, i.e., the asymptotic tracking problem of nD systems in the practical sense that not more than one of the variables of the system take infinite value simultaneously. A definition of skew primeness on the ring of practical stable rational function is given and corresponding solution conditions and procedures are presented. Based on these results, the practical tracking problem is solved and its application to the design of iterative learning control system is shown.

抄録全体を表示

Systems having symmetric transfer matrices often occur in applications. In this paper, digital control problems for continuous-time plants having symmetric transfer matrices are considered. The following results are obtained: 1) Symmetry of a transfer matrix of a continuous-time system is preserved through a discretization with generalized sample and hold functions under a mild assumption. 2) If a sampled-data H_∞ control problem has a solution for a given sampled-data generalized plant composed of a continuous-time 4-port system having symmetric transfer matrix, an integral sampler, and a zero-order hold, then there exists a sampled-data H_∞ controller having symmetric transfer matrix.

抄録全体を表示

In a factory automation, sequential control is one of the most important tequniques. In a sequential control system, the simulation of control program and the fault diagnosis of machine are very useful. In particular, for fault diagnosis, we must consider not only quick detection of fault parts but also continuation of operation. This is fault-tolerant technique. The achivement of high availability of a controlled system is important.
We reserch a useful fault diagnosis method of sequential control system. First, we construct Petri nets for control and controlled system, and simulate the movement of sequential control system with the Petri nets. Next, using the simulation model with Petri net, we carry out the fault diagnosis of the real sequential control system. For input signal of Petri nets for control, we use two input signals transmitted from Petri nets for controlled system and from real controlled system. Thus, by being duplication of input signal of Petri nets for control, this fault diagnosis method have so advantages as prevention of fault spread and fault-tolerance.
In this paper, we describe the new fault diagnosis method of sequential control systems, the realization of fault-tolerance with Petri net and the simulation of the sequential control system using the method of a vehicle moving between the yards in a factory.

抄録全体を表示

A crane active on the construction site is the loading machine that transports load to a fixed position by doing the turn movement centering on a part of the boom. The fulcrum of the rope draws circular arc tracks in such turn cranes and the behavior of load is very complex. Therefor the operations that maintain sway of a turn crane depend on an operator's skill. We propose anti-sway control method of the rotary crane based on a skillful operator's knowledge. The computer simulation and experiment of the loading control were executed by using this method. It was confirmed that the proposed method is effective.

抄録全体を表示

Universal Learning Network (ULN) which can be used to model and control the large-scale complicated systems in the network framework has been presented. In this paper, a method to construct the fuzzy model with a multi-dimension input membership function using ULN is presented. The fuzzy model under the framework of ULN is called Universal Learning Network-based Fuzzy Inference System (ULNFIS), which possesses certain advantages over other networks such as neural networks. It is also introduced how to imitate a real system with ULN and how to construct a control scheme using ULNFIS. ULN allows any differentiable non-linear functions as node functions. And simulations are carried out in order to compare the performance of the fuzzy control using ULNFIS with the neural network control. It is shown that the fuzzy control has better performance for the generalization capability than the neural network control.

抄録全体を表示

For successful applications of the genetic algorithm, there are two important points to be considered. The first point is the design of the fitness landscape introduced by the representation of the solution as a chromosome and searching operations such as crossover and mutation. The second is control of the convergence brought about by the selection operation. In the conventional implementation of GA, these two points are mutually dependent, i.e., a suitable selection pressure varies largely depending on, e.g., the crossover operator. Hence, it requires much trial-and-error effort to find a nice configuration of GA. In the present paper, the authors apply a novel selection rule, the Thermodynamical Genetic Algorithm (TDGA) proposed by Mori et al. to the traveling salesman problem (TSP), and propose an adaptive annealing schedule of the temperature in TDGA. Computer simulation with several crossover operators for TSP shows that TDGA reduces the mutual dependency between the fitness landscape and the convergence process. Furthermore, TDGA is applied to a hybrid algorithm which combines GA with the 2-opt method, a local search technique. The effectiveness of the TDGA in the hybrid approach is also confirmed through a comparison with other methods for solving the TSP.

抄録全体を表示

Current system identifications require nonlinear modeling technology. The system to be modeled has been effected by disturbances and has been related to human nonlinear actions. For this purpose, the black box modeling technology such as neural network is useful. However, the black box models can not show the qualitative relation between inputs and output. Therefore, the black box models require complete data set for modeling. In the real world, the complete data set seldom be obtained from the actual system. In spite of the data are incomplete, the traditional black box modeling techniques convert the information of the incomplete historical data to the model parameters which regulate the relations between input space and output space. These specified model parameters can not evaluate the reliability of the output for new input situations. This is a barrier to the actual use of traditional black box model for the real world system.
In this paper the nonlinear modeling technology using historical data for case is proposed. This technology is named TCBM (Topological Case Based Modeling). TCBM employs case based reasoning (CBR) technology for modeling to cope with the problems of traditional modeling technology that is a reliability of a estimated output on condition that the incomplete system data are used for modeling. CBR theory provides no general method to define the similarity between cases. In contrast, TCBM can generally define the similarity on the condition that the system to be modeled satisfies the continuous relationship between inputs and output. As a result, TCBM can easily construct its model same as traditional black box modeling technology and can show the reliability of the estimated output.

抄録全体を表示

State-space digital filters can synthesize the optimal filter structure with respect to overall quantization effects, but have many more multiplications than a canonical structure has. Therefore the sampling rate and the latency become important problems on implementing these filters. Especially for the very long latency the control system becomes unstable when these filters are used in the feedback control systems, such as the robot control system.
Previously we proposed a VLSI-oriented highly parallel architecture for state-space digital filters with high sampling rate and small latency. For the purpose of further speeding up and reducing hardware complexity, the distributed arithmetic, of which processing time depends on only word length, was applied to this architecture, making good use of high accuracy of state-space digital filters. Therefore, the very high sampling rate can be implemented independently of the filter order and the numbers of the inputs and outputs.
This paper designs and evaluates high-performance VLSI processor based on our proposed VLSI architecture for state-space digital filters using distributed arithmetic. As a result, we confirm that the processor can perform the sampling rate of about 1.67-MHz and the latency of 600ns using 0.8μm CMOS technology, in the case where the filter order is 16, the numbers of inputs and outputs are 1, respectively, and the word length is 14. Moreover, we show that the high sampling rate of about 1.6-MHz can be held constant, even if the filter order and the numbers of inputs and outputs increase.

抄録全体を表示

The difference image between the background image and the observing image is generally applied to a car tracking. But in the outside scene, the background image must be renewed on demand because the weather and the lighting condition change hourly. And long shadows, in the morning and evening, cause a miss extracting.
Therefore, we propose a new method for making an adaptive background image which enables to extract just bodies of cars by only to get a difference image between the background image and the observing image. Two images are required to make such a background image. One is the image-1 which consists of only static objects, while moving objects are replaced with its backing static objects, and the other is the image-2 which includes shadows of cars. The background image is composed by adding these two images. The difference between this new background image and observing image only contains the required moving objects.
Concerning the algorithm, any image-1 pixels show the gray level which kept the same level for some frames. If it turns to a pixel of moving objects, this method still keeps the pixel of previous static object. The gray level of image-2 is estimated by extracting the area of shadows. To extract these area, flood-fill algorithm is used and it makes possible to be robust against a flexible shape and its disappearance.
The effectiveness of this method is demonstrated by applying it to real images of a parking area and feeding the result to a car tracking program. In this experiment, miss recognition of moving object is remarkably restricted on any occasion. This method is not based on model matching, so it is useful and causes good result on any cases of extracting moving objects.

抄録全体を表示

A new stochastic evaluation method of objective sound level fluctuation under a contamination of background noise is proposed through the vibration level measurement on the basis of a generalized regerssion analysis between sound and vibration level fluctuations.

抄録全体を表示

We have developed an emissivity measurement system by which we can measure changes of spectral emissivities of metals under some conditions such as vacuum, oxidation and deoxidization. This system is an effective tool to enable us to formulate the behaviour of emissivities.

抄録全体を表示

A method for synchronous position control of mechatronic servo system using the first order model was extended to the method by use of the second order model. The proposed extended method was effective in the operation of higher speed conditions. The robustness of the method for the modelling error was analysed and proved. The proposed method can be effectively applicable in an industrial mechatronic servo systems.

抄録全体を表示

This paper describes an estimation method of fractal dimension. We propose Center of Moving Average method that is our new estimation method of the fractal dimension for time series data. We analyze the fractal dimension for stock prices by using this method

抄録全体を表示