巻号一覧
39 巻, 6 号
選択された号の論文の12件中1～12を表示しています
- |<
- <
- 1
- >
- >|
-
竹茂 求, 那須 潜思, 佐々木 正明, 高田 稔, 鹿股 昭雄, 小川 廣幸2003 年 39 巻 6 号 p. 521-527
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーAs an improved methodology for rapidly enumerating viable bacteria, microbial colonies in agar media are observed in the microscopic fashion at regular rate from the initial stage of culturing while media plate is being incubated. At each sampling time, in order to observe colonies in various depths without focusing on any of them, an agar plate is irradiated by a point light source to project the clear shadow of colonies onto the CCD image sensor. The projected image is processed with a computer to detect the colonies automatically. As the result of the experiment with E. coli, the number of colonies had reached to the plateau after six hours of incubation and the enumeration process was completed.抄録全体を表示PDF形式でダウンロード (3075K) -
馬場 充, 大谷 幸三, 礒田 祐輔2003 年 39 巻 6 号 p. 528-536
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーThe 3-D shape of an object is the most important measurement information data for many applications. Among numerous 3-D measurement method developed, a triangulation light-stripe rangefinding has proven reliable and is most commonly employed. The light-stripe rangefinder is usually required to acquire range data in a short time. Thus far, however, it has been difficult to increase the rate of conventional light-stripe rangefinders because most existing models make use of mechanical scanning mechanisms, such as a rotating mirror. Thus, to realize the goal of a very fast rangefinder, it is necessary to consider improvements of the scanning mechanisms themselves. Although rangefinders with a non-mechanical scanning mechanism were developed to avoid mechanical scanning, they have the disadvantages of low spacial resolution. We here propose a new, fast rangefinding scanning mechanism based on a new design, which, in contrast to conventional mechanical and non-mechanical scanning mechanisms, meet the criteria of fast scanning time and high spacial resolution. The proposed scanning mechanism works as follows: (1) First, we set the intensity of three stripes uniformly and project on the object; (2) second, by adjusting the intensity of the three light-stripes, a combined parabolic output with a single peak is generated; and (3) when the intensity of one of the light sources is lowered, the peak position of the combined output moves to the side of lower intensity. We have designed and built a prototype rangefinder able to acquire a three-dimensional position in 20ms per one point, which outperforms conventional rangefinders. We conclude that the proposed method produces a fast and effective rangefinder to be implemented in many applications.抄録全体を表示PDF形式でダウンロード (4314K) -
〓 明〓, 五味 裕章2003 年 39 巻 6 号 p. 537-543
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper describes a robust realtime estimation method of time varying human multijoint arm viscoelasticity during movement. The general idea of the present method will be described as follows: 1) The time varying human multijoint arm dynamics is modeled by two factors, simplified musculoskeletal dynamics and uncertainty factor consisting signal noises and modeling error of a rigid body dynamics. Additionally, pseudo-random perturbation is supplied so that its frequency beyond that of the voluntary component of the movement. Utilizing this feature, perturbed component of kinetic data can be extracted by band-pass filter. 2) By using the proposed model, a robust Kalman filter is designed for identifying the multijoint viscoelasticity of the human arm. The uncertainty factor can be considered by using some aspects of prior information of the noise and error. The proposed method is evaluated by demonstrating several simulation results of estimating arm viscoelasticity during movements.抄録全体を表示PDF形式でダウンロード (1469K) -
増淵 泉2003 年 39 巻 6 号 p. 544-551
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper is concerned with stabilization of implicit systems via interconnection. Stability of implicit systems is defined so that every solution in the sense of Impulsive-Smooth Behavior is impulseless and convergent. Preservation of the initial values of the implicit system (initial-freedom preservation) through interconnection is adopted as a formulation of regularity of interconnection. In order to analyze stabilization with initial-freedom preservation, we define notions of complete-stabilizability and zero-detectability that generalize stabilizability and detectability of state space systems. It is shown that stabilizability via, interconnection with initial-freedom preservation is equivalent to complete-stabilizability and zero-detectability. Thus standard results on output feedback stabilization of state-space systems are extended to stabilization of implicit systems via interconnection. Further, we provide criteria for each of complete-stabilizability and zero-detectability in terms of the Kronecker form, Lyapunov equation and inequality, rank and invertibility of the system pencil.抄録全体を表示PDF形式でダウンロード (1605K) -
小菅 一弘, 松本 大志, 盛永 真也2003 年 39 巻 6 号 p. 552-558
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, we propose a dynamic collision detection system using a nonlinear adaptive control scheme proposed by Slotine and Li. The system detects collisions of a manipulator with its environment without using external sensors. An adaptive control scheme is employed to control the manipulator and to estimate its dynamic parameters. The system detects collisions based on the difference between the actual input torque to the manipulator and the reference input torque, which is calculated based on the estimated parameters of the manipulator dynamics. The proposed collision detection scheme is applied to an industrial manipulator and experimental results illustrate the validity of the proposed scheme.抄録全体を表示PDF形式でダウンロード (3177K) -
向谷 博明2003 年 39 巻 6 号 p. 559-568
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, Nash strategy for multimodeling systems is considered. Compared with the existing results, it is shown that the near-optimal Nash strategy is given without non-singularlity assumptions for the fast subsystems. Furthermore, the asymptotic structure of the cross-coupled algebraic Riccati equations are established. Finally, a new near-optimal Nash strategy is obtained which does not depend on the values of the small parameters. It is proven that the near-optimal strategy can achieve a performance which is O(ε) close to the optimal performance.抄録全体を表示PDF形式でダウンロード (1954K) -
Daisuke YASUFUKU, Yuji WAKASA, Yutaka YAMAMOTO2003 年 39 巻 6 号 p. 569-574
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper proposes an adaptive digital filtering method which improves continuous-time performance. To take account of continuous-time performance, we make an assumption that the adaptive filter can exploit the continuous-time error signal, and define the integral of the square error signal as a performance index. Using lifting, the performance index can be equivalently reduced to a summation of discrete-time signals. Therefore it is shown that the recursive formula for the filtering problem with the performance index is given by the well-known RLS algorithm. Next, a more implementable case is considered where the output of the unknown system is approximated by the oversampled discrete-time signal and the summation of the squared discrete-time signals is defined as an approximate performance index. It is shown that the approximate performance index can be directly applied to the first result. Finally, some numerical examples are given to illustrate the effectiveness of the proposed method.抄録全体を表示PDF形式でダウンロード (1467K) -
山口 博明, 新井 民夫2003 年 39 巻 6 号 p. 575-584
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a path following feedback control method for a cooperative transportation system with two car-like mobile robots coupled together by a carrier. We first define seven variables for describing the state of the system. These variables are the relative position counted as two and relative orientation of the first robot to a path followed, the first robot steering angle, the relative orientation of the carrier to the path, the relative orientation of the second robot to the path, and the second robot steering angle. We secondly transform the system into two-chain, single-generator chained form, based on Frobenius' theorem in differential geometry. This means that we make such transformation in a coordinate system where the path is a curved line axis and a straight line perpendicular to the tangent of the path is another axis. In other words, this conversion is more general because it includes a conversion performed in the Cartesian coordinate system where its two axes are straight lines and they are perpendicular to each other. One of the seven variables of the form is the moving distance of the first robot along the path. By moving the cooperative transportation system, the presented feedback control method makes the other six variables converge into desired ones, which is equivalent to making the system follow the path. Especially, the system can follow any path as long as its curvature is two times differentiable. We have performed a path following simulation where the first robot follows a circle path and the relative orientation of the carrier to the path converges into a specified angle. The validity of the transformation and the feedback control method is supported by computer simulations.抄録全体を表示PDF形式でダウンロード (2175K) -
平田 研二, 太田 快人2003 年 39 巻 6 号 p. 585-589
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーThis paper considers an approximation problem of the state reachable set for linear discrete-time systems with unknown but bounded disturbance inputs. A simple recursive procedure provides a sequence of enlarging convex polyhedral sets which successively approximate the state reachable set. The notion of positive invariance is utilized to derive the stopping criterion for the proposed recursion. The procedure makes possible that an approximation of the state reachable set within the required accuracy.抄録全体を表示PDF形式でダウンロード (1202K) -
岡村 寛之, 石倉 武, 土肥 正2003 年 39 巻 6 号 p. 590-599
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーDynamic power management (DPM) is one of the most effective techniques for reducing energy consumption. Especially, auto-sleep function is the simplest but effective way to reduce electric power consumption. In this paper, we consider a stochastic model to determine an auto-sleep timing sequentially. More precisely, we develop an optimal control scheme of auto-sleep timing based on the Q-learning, which is a part of reinforcement learning algorithms and is strictly related to the Markov decision process (MDP) and the semi-Markov decision process (SMDP). First, we reformulate the stochastic auto-sleep model under the SMDP. Second, the optimal control scheme is to determine the optimal auto-sleep timing established by applying the Q-learning algorithm. Finaly, numerical examples are presented to investigate the effectiveness of DPM with real data.抄録全体を表示PDF形式でダウンロード (1858K) -
酒井 正夫, 本間 経康, 阿部 健一2003 年 39 巻 6 号 p. 600-606
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーIn this paper, a statistical approximation learning (SAL) method is proposed for training a new type of neural networks which is known as simultaneous recurrent networks (SRNs). SRNs have the capability to approximate non-smooth functions which cannot be approximated by using conventional feedforward neural networks. However, most of learning methods for SRNs are computationally expensive due to their inherent recursive calculations. To solve this problem, a statistical relation between the time-series of the network outputs and the network configuration parameters is used in the proposed SAL method. Simulation results show that the SAL method can learn a strongly nonlinear function efficiently.抄録全体を表示PDF形式でダウンロード (1613K) -
杉山 昭暢, 小谷 学2003 年 39 巻 6 号 p. 607-613
発行日: 2003/06/30
公開日: 2009/03/27
ジャーナル フリーDNA microarray technology has facilitated in obtaining massive gene expression data. Efficient categorizing methods are necessary to analyze such massive data. There are several clustering methods to categorize the gene expression data into functionally meaningful groups. One method is a Self-Organizing Map (SOM) that presents high-dimensional data by low-dimensional data. However, it is difficult to find clustering boundaries from results of the SOM. To understand how the SOM categorizes the gene expression data, we apply a method of combining the SOM and a clustering method based on minimization of within-class covariance matrix. We show that the proposed method is effective for categorizing the published data of yeast gene expression.抄録全体を表示PDF形式でダウンロード (1725K)
- |<
- <
- 1
- >
- >|