計測自動制御学会論文集
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
37 巻, 11 号
選択された号の論文の14件中1~14を表示しています
  • 呉 景龍, 水原 啓暁, 根来 清, 橋田 昌弘, 小笠原 淳一, 山内 秀一, 松永 尚文, 斉藤 俊
    2001 年 37 巻 11 号 p. 1003-1011
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    Human visual and auditory systems have been evolving through the long history of human being and have attained very high performance. However, artificial systems for those are often poorer in performance despite of remarkable progress in their efficiency. In order to construct artificial systems with high performance such as human visual and auditory systems, it is necessary to elucidate the mechanisms of human visual and auditory systems. In order to investigate the characteristics of human visual and auditory advanced information processing, the authors have measured the advanced function of human brain related with the calculation processing. In this study, to construct the tasks measuring the ability of human calculation information processing, the characteristics of the visual and the auditory calculation information processing were measured with using the mental arithmetic problems as the visual and the auditory stimuli by psychological experiment. Furthermore, activated areas of human brain related with the calculation processing were also measured by functional magnetic resonance imaging (fMRI). Based on the measurement results, the authors discuss the difference between the visual and the auditory calculation information processing. The results of psychological experiment and fMRI will be significant the mechanism of visual and auditory information processing and to construct artificial systems.
  • 香川 昌己, 花泉 弘
    2001 年 37 巻 11 号 p. 1012-1019
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    This paper describes an InSAR (Interferometric SAR) processing system which we developed for obtaining accurate interferogram, Digital Elevation Model (DEM) and Differential Interferogram. There have been proposed many methods for deriving these products from SAR raw data. In the most of former methods, the co-registration was made using a linear function of the range pixel and the azimuth line. The linear function is, however, insufficient to represent local disparities around peaks or valleys. In the proposed system, two Single Look Complex (SLC) images to be co-registered are divided into many triangles so that vertices of these triangles are the corresponding points obtained by the offset measurement process. After the division, images are co-registered by applying Affine transformation to each triangle pair. The piecewise co-registration removes disparity at least on the corresponding points. On the other hand, only the nominal value of the baseline length is currently provided for the interferometric processing. The fixed value, however, does not represent non-parallel orbits from which raw interferogram is obtained. The ambiguity causes the geometrical distortion in DEM and the apparent displacement in the result of differential InSAR processing. The system derives coefficients of both functions for baseline length and for inclination angle by fitting the local disparities to a geometrical model.
    Using the system, the accuracies of both coherence image and interferogram derived from actual SAR raw data were successfully improved. The baseline estimated by the system reduced apparent displacement in differential interferogram to 1/10 of that yielded by the nominal baseline.
  • 浦久保 孝光, 土屋 和雄, 辻田 勝吉
    2001 年 37 巻 11 号 p. 1020-1025
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, we deal with the control problem of a class of non-holonomic systems with m input variables. In the system, if the input vector fields and the first level of Lie brackets between them span the tangent space at each configuration, the system can move in every direction using the first level of Lie bracket motions and it is controllable. This class of systems is called as first order systems. We derive a time-invariant discontinuous state feedback law for the system based on the Lyapunov control: The input is designed so that the derivative of the Lyapunov function is composed of a symmetric and an asymmetric bilinear form in the gradient vectors of the Lyapunov function. In the controlled system, the desired point is the only stable equilibrium point. The performance of the controller is verified by numerical simulations.
  • 池田 貴幸, 深谷 正和, 美多 勉
    2001 年 37 巻 11 号 p. 1026-1033
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    We have proposed a new control method called variable constraint control which uses the first integral of constraints as an invariant manifold obtained by adding controlled constraints to nonholonomic mechanisms. We apply results to position and posture control of an underwater vehicle and investigate relations between control singularities and initial posture of the vehicle.
  • 深尾 隆則, 山脇 明, 足立 紀彦
    2001 年 37 巻 11 号 p. 1034-1039
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    The control technique which combines H control with adaptive nonlinear control is applied to the active suspension system which is divided into two parts: a main car body part and a hydraulic actuator part. The H control design is used for the body part where the frequency response property is important to improve the ride quality of the automobiles, and the adaptive nonlinear control design is used for the actuator part with the high nonlinearities and the parametric uncertainties which is required to transmit the power to the main body adequately because it gives large effects on the control result. These two design methods are combined by a modified version of backstepping which is a powerful tool for nonlinear control and adaptive control. Finally some simulation results are provided to confirm the usefulness of our method.
  • ステレオビジョンによる斜度推定とその斜面歩行への応用
    熊谷 正朗, 江村 超
    2001 年 37 巻 11 号 p. 1040-1047
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    The bipedal walk is a useful method for moving in human live environment. However, there are few research works on biped robots on uneven ground though adaptability to those environments is necessary for them. In this paper, the slope-walk of a biped robot based on stereo vision is described. First, a method to estimate inclination with vision is proposed. This method is based on triangulation. The robot observes the ground in front of it at a certain angle below horizon and measures distance to there with stereo vision. This distance becomes smaller on uphill slope and larger on downhill way. The vision is also used for target detection for followed-walk. The robot detects distance and direction of a target. We propose direct utilization of parallax as distant information and direct conversion from parallax to inclination. These methods provide enough accuracy for walking robot and reduce cost of calculations and calibrations. Secondly, a sensor based walk is proposed. The sequence of walk consists of two parts. One is stabilization phase that absorbs attitude-disturbances of robot caused by environment or alteration of gait. This stabilization uses attitude sensors mounted on the body. The other is stepping phase. We designed simple gaits so that it can be altered easily for slope-walk and curve-walk. The experiments using a biped robot with legs of human type degree of freedoms and a stage that provides various slopes were carried out. The accuracy of slope estimation was about ±1[deg] and the walk through undulating ground of 5[deg] downhill to 5[deg] uphill way was achieved. The results showed that our method gains adaptability of biped robots to slopes.
  • 柳井 法貴, 山本 元司, 毛利 彰
    2001 年 37 巻 11 号 p. 1048-1055
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    This paper discusses control method for crane using inverse dynamics calculation of the system. The load of crane is generally easy to swing and thus the work using the crane is dangerous. For preventing the swing of the load, it is desirable to feedback the position of the load in a control method. As such a control law, state-feedback control based on linearization of the crane system is often used. However the method reduces control performance when rope length is changed, because of non-linearity of the system. In this paper, a new feedback control method is proposed where the inverse dynamics calculation is involved to compensate non-linear dynamics of the system. In the control system, an observer is used to get higher order derivatives of the position of load, which is needed for the inverse dynamics calculation. By using the control method, control performance is unchangeable for changing rope length. Numerical simulation and experiment are shown to verify the effectiveness of the proposed control method.
  • 芳谷 直治
    2001 年 37 巻 11 号 p. 1056-1061
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    In pole-placement control, offset removal techniques against unknown disturbances containing model errors are classified into two methods: (I) disturbance compensation method, and (II) integration method. Most of the previous studies recommend method (II). In this paper, the author compares both methods and shows that method (I) is comprehensive, including method (II) as the simplest case. Compared with method (II), method (I) can realize smaller input/output variances and larger robustness to model errors by designing the filter for disturbances (the disturbance observer). Hence, an optimal design procedure for the filter is proposed to take into account the estimated magnitudes and the characteristics of disturbances and model errors. A numerical simulation illustrates the advantages of method (I).
  • 中村 政俊, 国松 穣, 後藤 聡, 久良 修郭
    2001 年 37 巻 11 号 p. 1062-1068
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    Contour control of robot arms is required high speed and high accuracy performance in industries. This paper describes a method of contour control of the industrial articulated robot arm by use of the synchronous positioning control with dynamic compensation. The proposed method compensates for master and slave dynamics delay with the modified taught data method based on the second order model, and enables accurate contour control at high speed. Effectiveness of the proposed method was assured by the experiment of an actual industrial robot arm.
  • 佐藤 剣, 片井 修, 堀内 匡, 井田 正明, 川上 浩司
    2001 年 37 巻 11 号 p. 1069-1077
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    A way of acquiring fuzzy inference rules is introduced based on the novel notion of fuzziness referred to as constraint-interval fuzzy set. The basic apparatus of this rule acquisition are a method of generalizing instance data (instance information) and also a method to organize and to refine these generalized pieces of instance information. For the instance generalization, we will introduce a new method utilizing pairs of instance data, which we call DIG (Double Instance Generalization). This method is based on the notion of convexity of predicates. For the organization and refinements of generalized pieces of information, we will employ a Hopfield type neural network, by which effective ways of extracting fuzzy inference rules from a collection of instance data on fuzzy constraints or crisp constraints are derived.
  • 人との関わりを重視したシステムの設計に向けて
    須藤 秀紹, 川上 浩司, 片井 修
    2001 年 37 巻 11 号 p. 1078-1086
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, we will propose a representational model of “interactive” artifacts which reflects interactions between human and artifacts base on the discussion of alethic, deontic and temporal aspects of the interaction. From these three types of modality, we will define three layers to represent the interactions. The base layer represents causal relations which are governed by physical laws or effects. The main layer represents state transitions by unrestricted operations. Some operations are restricted by teleological necessities which are derived from designers' intention. The top layer represent this type of restrictions by using what we call “task unit graph”. The tight interactions among the three layers explain interactions among designers, operator and environment via an artifact.
  • 三宅 美博, 宮川 透, 田村 寧健
    2001 年 37 巻 11 号 p. 1087-1096
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    Human-human system can coordinately self-organize relational functions by their mutual communications. In this study, we aimed to establish a design principle of man-machine system to realize such a co-emergence process in human-human interaction. This developmental dynamics is modeled based on duality of human communication which is composed of simultaneous process in open system and causal relationship in closed system. The former is represented by body model and is to organize coherence between human and machine, and the latter is internal model to separate the coherence into two one-sided actions. Through mutual constraint between these two sub-dynamics, relational functions as a whole system is self-organized in the man-machine interaction. Using this duality model, co-emergence process was realized as a mutual-adaptation between human and machine, and its effectiveness was shown in a walk support robot.
  • 竹前 嘉修, 斎藤 英雄, 小沢 慎治
    2001 年 37 巻 11 号 p. 1097-1103
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, we propose an automatic evaluating system of human skin surface condition based on subjective evaluation provided by cosmeticians. In the proposed system, image features extracted on skin image taken by the CCD camera and subjective evaluation by cosmeticians are flexibly connected by using a back-propagation neural network, so that it can automatically estimate human skin surface condition based on subjective evaluation from our various skin images. Using the trained neural network, human skin surface condition based on subjective evaluation is estimated for unlearned skin images with real time processing. Then subjective evaluation by this system was compared with that by cosmeticians. Since the proposed system can successfully estimate human skin surface condition like cosmeticians, the effect of the system is demonstrated.
  • 西部 優奈, 呉 海元, 塩山 忠義
    2001 年 37 巻 11 号 p. 1104-1110
    発行日: 2001/11/30
    公開日: 2009/03/27
    ジャーナル フリー
    When the blind walks using a white cane, it is very difficult to detect down steps. For the purpose of mobility aid for the blind, this paper presents a new method for detecting the position and direction of train platform curb which is one of the most dangerous down step, by using an image sensor. First, the lines of lights on the ceiling, and horizontal lines of the station signboard are extracted from an image. Then, the candidate lines of the train platform curb are extracted, and the true train platform curb is determined by selecting the candidate lines with the information of the lines of lights and horizontal lines of the station signboard. The effectiveness of the proposed method is shown from extensive experiments for real images of subway platform scenes.
feedback
Top