システム制御情報学会論文誌
Online ISSN : 2185-811X
Print ISSN : 1342-5668
ISSN-L : 1342-5668
29 巻, 5 号
選択された号の論文の6件中1~6を表示しています
論文
  • Morita Kento, Kobashi Syoji, Wakata Yuki, Ando Kumiko, Morimoto Masaka ...
    2016 年 29 巻 5 号 p. 195-201
    発行日: 2016/05/10
    公開日: 2016/08/10
    ジャーナル フリー
    Magnetic resonance (MR) images are widely used to diagnose cerebral diseases. MR image registration based brain shape evaluation is effective to diagnose brain diseases because the diseases may deform the brain shape, and the deformed region differs among types of diseases. This paper focuses on neonatal brain MR images, and introduces a non-rigid image registration method using sulcal-distribution index (SDI), which is calculated from MR signal on the cerebral surface. The method is based on an iterative closest point (ICP) algorithm. The proposed method will be effective for the neonatal brain because the method evaluates the correspondence of cerebral sulci distribution. Results in seven neonates (modified age was between 3 weeks and 2 years) showed that the method registered the cerebral shapes successfully.
  • 木下 拓矢, 井上 明法, 大西 義浩, 山本 透
    2016 年 29 巻 5 号 p. 202-209
    発行日: 2016/05/10
    公開日: 2016/08/10
    ジャーナル フリー
    In industrial processes, it is necessary to maintain the user-specified control performance in order to achieve desired productivity. This paper describes a new design scheme of performance-driven controllers in which `controller design'’ is fuzed with `control performance assessment’'. According to the proposed scheme, control parameters are tuned by using a FRIT (Fictitious Reference Iterative Tuning). Therefore, the control performance is improved by only once using closed-loop data. The proposed scheme is a kind of data-oriented controllers whose control parameters are directly calculated without estimating system parameters. The effectiveness of the proposed method is verified by using a simulation example and experiment results.
  • 信川   創, 西村  治彦, 山西  輝也, 劉   健勤
    2016 年 29 巻 5 号 p. 210-215
    発行日: 2016/05/10
    公開日: 2016/08/10
    ジャーナル フリー
    Izhikevich neuron model, which combines continuous spike-generation mechanisms and discontinuous resetting process after spiking, can reproduce almost all spiking activities including chaotic spiking in actual neural systems. When the chaotic state is evaluated in this model, it is known that conventional Lyapunov exponent where the continuous trajectory is presupposed cannot be applied due to the state dependent jump in the resetting process. To evaluate Lyapunov exponent in the system with the resetting process, the accurate numerical calculation for the trajectory by Newton method and the consideration for saltation matrix are needed. By virtue of this method, several routes to chaos have been found in Izhikevich neuron model. While on the other hand, in this study,we have proposed the method combining Euler method and Lyapunov exponent on Poincaré section and evaluated the chaotic state in Izhikevich neuron model. As the result, it has been confirmed that this method can also judge the chaotic state by tuning the initial perturbation against the trajectory.
  • 英   慎平, 中迫   昇, 中山  雅人, 篠原  寿広, 上保  徹志
    2016 年 29 巻 5 号 p. 216-224
    発行日: 2016/05/10
    公開日: 2016/08/10
    ジャーナル フリー
    In many engineering fields, it is very important to measure the distance and velocity of a moving target. In order to measure the distance to a stationary target, we previously proposed an Acoustic Distance Measurement (ADM) method that is based on the interference between the transmitted and reflected waves. However, due to the Doppler effect, that ADM method results in measurement errors when the target is moving. In this paper, we propose a distance and velocity estimation method for targets moving at a constant velocity; it can be used at short range, and it is based on the interference of the reflection of a single linear chirp. In particular, we introduce frame processing to the measurement of the distance to a moving target. That is, the velocity of the moving target is first estimated by considering the estimated distances in each short interval; the initial position and the distance to the moving target are then evaluated. We confirm the validity and effectiveness of the proposed method through computer simulations and experiments.
  • 森田  亮介
    2016 年 29 巻 5 号 p. 225-230
    発行日: 2016/05/10
    公開日: 2016/08/10
    ジャーナル フリー
    Subtractive dither method is that artificial noise called “dither signal” added to the original signal before quantization and subtract the same noise from the quantized signal. In this paper, the method is applied into quantized feedback control systems and the performance of quantizers with subtractive dither is analyzed. One of the analyzed quantizer is that using usual subtractive dither. In addition, another type of the quantizer which is using “asymmetric subtractive dither” is also analyzed.
研究速報
feedback
Top