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, 2 号
選択された号の論文の10件中1~10を表示しています
  • 水本 郁朗, 岩井 善太, 小原 浩司朗
    2001 年 37 巻 2 号 p. 115-124
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    The property of systems that can be rendered exponentially passive via output feedback is called output feedback exponential passivity. The design of adaptive output feedback control for output feedback exponentially passive MIMO nonlinear systems is proposed. Further, the alleviation of the output feedback exponentially passive conditions by introducing a parallel feedforward compensator (PFC) will be considered and design principle for the PFC is derived. The new concrete design scheme for the PFC is also given.
  • 芳谷 直治
    2001 年 37 巻 2 号 p. 125-132
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, we propose a new algorithm of dual predictive control for a system expressed by a linear autoregressive and exogenous (ARX) model with uncertain model parameters. The cost function, represented by the Bellman's equations, is based on the cost for the generalized predictive control (GPC). To avoid impractical computational burden, a new sub-optimal cost function which approximates the Bellman's equations is derived. The sub-optimal cost contains the variances of various future uncertainties, whose exact values are difficult to obtain. To cope with this, an algorithm to calculate the cost is devised with some appropriate approximations. The main difference from the preceding works on dual predictive control is the consideration of future uncertainties in control input and in the output values in the AR part of the model. Hence, the proposed control algorithm including the cost function is the closest to the optimal among all the dual predictive control algorithms. The control input to minimize the cost is obtained through nonlinear optimization. A numerical example illustrates the effectiveness of the algorithm.
  • 久保 智裕
    2001 年 37 巻 2 号 p. 133-138
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    An alternative method to construct an optimal memoryless regulator of systems with time-delay in the states is proposed. A feedback law is constructed with a solution of a linear matrix inequality. It is shown that the resulting closed loop system satisfies the circle condition, and it is assured to have a good robustness property as well as finite dimensional linear quadratic regulators. Numerical examples are given.
  • 梶原 秀一, 橋本 幸男, 土谷 武士
    2001 年 37 巻 2 号 p. 139-145
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    Periodic stimulation sometimes stabilizes mechanical systems dynamically. we have investigated a swing-up control of an inverted pendulum from a view point of such a dynamic stabilization. First, we analyze the timing of the periodic input in which the pendulum is excited. Next, we propose the control method using a entrainment so that the periodic input is synchronized with the swing of the pendulum. The control input is consisted of a periodic solution of van der Pol's equation which is inputted the angular verocity of the pendulum. Finally we present experimental results that our controller can realize the selective swing-up control for a rotational double pendulum.
  • 牛島 淳博, 池田 雅夫, 村松 鋭一
    2001 年 37 巻 2 号 p. 146-151
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    Modeling of a plant with variable operating conditions is considered. The plant is described by an interpolation of proper stable coprime factorizations of two transfer functions obtained at representative operating points. Nonlinear interpolation functions are introduced in the numerator and denominator of the interpolated model individually, which tune the interpolation parameters at reference operating points. A problem of finding suitable nonlinear interpolation functions and coprime factorizations of the transfer functions at representative operating points is formulated so that the nonlinearly interpolated model becomes the best approximation of the plant. A solution method is presented in the state space utilizing matrix inequalities.
  • 渡辺 一郎, 河辺 享之, 大塚 伸一, 小林 弘樹, 市原 順一
    2001 年 37 巻 2 号 p. 152-159
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    The authors have developed a low-cost, high-performance magneto-optical (MO) disk drive whose track-following mechanism is implemented as a linear positioner driven by a voice coil motor (VCM), that is, a single-stage tracking mechanism. In this tracking mechanism, Coulomb friction disturbance caused by the slide bearings of the positioner is a critical problem preventing precise track following.
    To solve this problem, we developed a learning controller, which automatically estimates and generates an ideal VCM current signal to perfectly compensate for the friction disturbance. The estimated signal pattern is approximated as the weighted sum of time-shifted rectangular functions, and the weight for each rectangular function is updated online using a simple learning algorithm during track following control. Simultaneously, the latest updated signal is added to the output of a feedback controller and fed to the VCM driver.
    We evaluated the efficiency of our controller both in a simulation and an experiment. In the experiment, the tracking error peaks caused by Coulomb friction were drastically lowered by using the learning controller, and a tracking error of less than 0.02μm was confirmed during operation of an MO disk rotating at 4, 500rpm with a radial track runout of 77μm peak to peak.
  • 加納 学, 大野 弘, 長谷部 伸治, 橋本 伊織
    2001 年 37 巻 2 号 p. 160-167
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In order to improve the performance of multivariate statistical process control (MSPC), two advanced methods, known as moving principal component analysis (MPCA) and DISSIM, have been proposed. In MPCA, changes in the direction of each principal component or changes in the subspace spanned by several principal components are monitored. DISSIM is based on the idea that a change of operating condition can be detected by monitoring a distribution of process data, which reflects the corresponding operating condition. The dissimilarity index was introduced for quantitatively evaluating the difference between two data sets. In the present work, the advanced methods, i.e. MPCA and DISSIM, and the conventional method based on principal component analysis were compared with their applications to the Tennessee Eastman process. The monitoring performance of MPCA is strongly affected by selection of principal components and time-window size. In particular, selection of principal components is crucial for effective functioning of MPCA. For a large process, monitoring a subspace spanned by several principal components can achieve better performance than monitoring each principal component. On the other hand, only a time-window size is a design parameter of DISSIM. From this viewpoint, DISSIM seems more practical than MPCA. The application results have clearly shown the superiority of MPCA and DISSIM over the conventional method. The advantage of MPCA and DISSIM comes from the fact that those methods focus on changes in the distribution of process data. Furthermore, a contribution of each process variable to the dessimilarity index is introduced for identifying the variables that contribute significantly to an out-of-control value of the index, and then the effectiveness of the proposed contribution is evaluated.
  • センサからモータまで
    柴田 克成, 岡部 洋一, 伊藤 宏司
    2001 年 37 巻 2 号 p. 168-177
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, Direct-Vision-Based Reinforcement Learning is proposed not only for the learning of motion but for the learning of the whole process, which includes recognition, from sensors to motors in robots. In this learning, raw visual sensory signals are put into a layered neural network directly, and the network is trained by Back Propagation using the training signal generated based on reinforcement learning. By employing neural network, whole the process becomes seamless and is trained purposively and harmonically.
    Two simulations of the mobile robot with visual sensors are performed as examples. One task requires stereo-vision, and the other requires obstacle avoidance. By observing the hidden representation in the neural network, it is shown that some abstract representation of spatial recognition is formed without any advance knowledge.
    Finally it is shown that each visual sensory cell makes a role of localization of the global information of the space and it helps the fast and stable learning.
  • 軸屋 一郎, 木村 英紀
    2001 年 37 巻 2 号 p. 178-180
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    We consider a model set in the chain-scattering framework. Given a rational transfer function matrix which we call a generator, a model set is defined by a feedback connection of the generator and the norm-bounded linear-time-invariant uncertainty. We obtain a necessary and sufficient condition on generators for inclusion and equivalence between model sets.
  • 石井 千春, 申 鉄龍
    2001 年 37 巻 2 号 p. 181-183
    発行日: 2001/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    This paper develops a tracking control problem of rigid-link electrically-driven robot manipulators with uncertainty. We extend a research line of the corrective robust tracking control scheme given by Dawson et al. (1992) to construct such a Lyapunov function that ensures not only stability of a tracking error system but also an L2-gain constraint for a tracking performance. A robust tracking controller with L2-gain disturbance attenuation is designed such that the closed-loop error system is globally stable in the sense of uniform ultimate bounded stability with the L2-gain less than any given small level.
feedback
Top