計測自動制御学会論文集
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
26 巻, 2 号
選択された号の論文の17件中1~17を表示しています
  • 葉 雄英, 藤村 貞夫, 山田 範秀
    1990 年 26 巻 2 号 p. 123-128
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, we describe a new method for determining 3D-shape of objects which consist of specular plane surfaces. The method uses a light source of grating pattern encoded by an M-sequence. Being reflected by a specular plane and perspectively projected into an image plane, the set of parallel source lines produce a set of image lines converging into a vanishing point. The orientation of the specular plane can be determined by the vanishing point and the slope of the source lines. The position of the plane can be determined by the correspondence between image lines and source lines which is obtained by using a characteristic regularity of M-sequence. The determination of surface planes allows us to reconstruct the specular objects.
    Simulation and experimental results are shown to verify the analytic expressions. The error in determining plane orientation and depth is discussed. The main advantage of the method is that it needs only one image.
  • 原田 博之, 柏木 濶, 高田 雅
    1990 年 26 巻 2 号 p. 129-136
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    A new method has been proposed for generating binary random sequence based on an M-sequence. The method is called random sampling method and the binary random sequence is generated by using successive k-tuples of an M-sequence and uniform random numbers.
    This paper describes the analysis for obtaining the optimal condition for the generating binary random sequence which has the most random property, by use of a new concept called merit factor Fr. The merit factor Fr is obtained from the expected values of the autocorrelation function of binary random sequence. The randomness of the binary random sequences generated by the random sampling method are evaluated by Fr. And it is shown that when the tuple length is equal to the period of the original M-sequence, the generated binary random sequence is considered to be the most random.
  • 松岡 陽一, 川瀬 成一郎
    1990 年 26 巻 2 号 p. 137-144
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    This paper proposes a single-satellite geolocation system to provide three dimensional position for moving vehicles. The elements of this system are single-geostationary satellite, three reference ground stations, control center (data processing center) and many users.The distinctive feature is that this satellite has 50m orthogonal crossed-booms to obtain long interferometry baselines.
    L-band signals transmitted by an user are received at each interferometry element. The signals are translated to C-band frequency and downlinked to a control center for data processing. The center keeps a constant watch over signals from three reference stations of known position for calculation of boom deflections including attitude errors and determination of user's position from measured phase differences and compensation data of boom deflections.
    In this paper, we discuss the problems of how to eliminate an ambiguity of measured path differences and to compensate for boom deflections. As a result of study, we have confirmed that a position determination accuracy of 200m to 300m is possible in this system.
  • 鳥越 一平, 石井 泰
    1990 年 26 巻 2 号 p. 145-150
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    The acoustic resistance (real part of the acoustic impedance) of an orifice increases in proportion to the mass velocity of the flow which grazes the orifice. The mechanism causing the flow dependence of orifice acoustic resistance-acoustic energy stored in the vicinity of an orifice is blown away by flow-is illustrated by a simple model of the flow above an orifice. A dimensional analysis was applied to this model and it has been found that five nondimensional parameters characterize the process.
    On reference to these theoretical investigations a trial acoustic impedance flowmeter has been built. Two orifices are drilled in the wall of an acoustic tube shaped like a “Pitot-static tube”. This acoustic tube is placed in such a position that the orifices are exposed to grazing flow, and is driven by a sound source at a constant velocity amplitude. The tube sound pressure at the orifices is picked up by a microphone. Phase-sensitive detector being employed, the output of the microphone is converted to a DC voltage, which is proportional to the orifice acoustic resistance and is the measure of the mass flow rate. Several measurements have been performed for air flow. And the mass flow rates measured with this flowmeter were in good agreement with the ones otherwise determined.
  • 大屋 勝敬, 西村 行雄, 米澤 洋
    1990 年 26 巻 2 号 p. 151-158
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper we propose a stabilizing control method of uncertain single-input, single-output systems by using a state observer. The controlled objects contain uncertain but bounded parameters and disturbances, and the nominal systems are described by output form.
    We consider the design method, in which the states are practical stable, on the coordinates transformed by a diagonal matrix T=diag[1, α-1, …, α-n+1] where α is positive and a design parameter. First, in the case where the driving vector of a nominal system is bTS=[0, …, 0, 1], it is shown that the control systems can be constructed in which the ultimate norm of the system output converges to any small region by increasing design parameter α. Next, in the case where the driving vector consists of some nonzero elements, it is shown that controlled object can be transformed to such a system as described in the first case by using a precompensator. Namely, the control system, having prescribed property, can be constructed. That precompensator is in proportion to the inverse function of the numerator polynomial. Therefore, the numerator polynomial must be asymptotically stable for guaranteeing internally stability of the control system.
    Finally, computer simulation results are presented to illustrate the effectiveness of the proposed method.
  • 郭 海蛟, 石原 正, 竹田 宏
    1990 年 26 巻 2 号 p. 159-166
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    We propose a design of discrete-time type 1 servosystems using output feedback with one-step computation delay. As a basic design, we consider an output feedback design using the state feedback design proposed by Mita and Mukaida and a filtering type Kalman filter. We show that this controller can be simplified by use of a prediction type Kalman filter without sacrificing achievable feedback properties. For the servosystem with the simplified controller, we obtain decomposed expressions for the sensitivity matrices at the input and output of the plant. These expressions contain the sensitivity matrix of the regulator or that of the Kalman filter used in the design and explicitly represent the effect of the integrator and the computation delay in the controller. Using these expressions, we show that the duality holds between the feedback properties at the input and those at the output of the plant. In addition, we discuss the sensitivity properties obtained by the well-known loop transfer recovery techniques. A numerical example is presented to illustrate the performance of the proposed design.
  • 沖田 豪, 小林 康秀, 田中 正吾
    1990 年 26 巻 2 号 p. 167-173
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    We propose the method to determine the system structure and to estimate the parameters for the nonlinear system, described by the functional expansion.
    First, we hypothesize candidate models of different structure. We may estimate the system parameters of each model by the proposed criterion function which is based on the maximum likelihood method. The estimated parameters are obtained analytically from noisy input and output data, and they are consistent.
    Secondly, we calculate a posteriori probability for each model by using Bayesian theorem. We adopt the model whose a posteriori probability is maximum as the optimal one.
    Finally, we verify the validity of this method by digital simulations.
  • 小林 史典, 木村 洋一, 渡邉 千次
    1990 年 26 巻 2 号 p. 174-179
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    Though ‘repetitive control’ can provide for significantly accurate operations, its performance in turn degrades dramatically when the reference period is not equal to the dead time in the repetitive controller. This paper proposes, bearing its application to path control in mind, an extension of repetitive control which can adapt itself to dynamical changes in the reference period.
    First, the dead time is driven by a PLL circuitry to statically track the reference period. The scheme is then refined by reducing transient errors by
    (1) switching the controllers, in which control variable waveforms are previously formed by enough number of repetitions, according to changes in the period,
    (2) forcing the state variables, with a feedforward operation, to fit itself to the new period.
    Also proposed is a practical scheme for cases where speed changes more than once a cycle. Since the cases cannot enjoy the effect of repetition, the compensator is thus excluded from the loop and repetitive control is made use of offline to form the control variable waveform.
    An application of the proposed method is then described: path control of a deburring robot which changes its path speed according to the cutting load. The experimental results show that the accuracy remains same as the case of constant speed.
  • 有川 晴彦, 廣田 薫
    1990 年 26 巻 2 号 p. 180-187
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    A new type of fuzzy inference chip adopting address-look-up and virtual paging method, i.e. sophisticated use of common digital memory devices, has been successfully developed. The characteristics of the inference chip are summarized first and then its theoretical background is introduced.
    This inference chip possessing robust noise immunity executes any kind of fuzzy operations and variables using a limitless number of fuzzy production rules with unlimited fuzzy linguistic labels and free-shaped membership functions in continuous valued mode at high-speed (more than 60 MFLIPS) and low-cost. It can be easily linked to the commonly used digital computer system and furthermore helps to pave the way for harmonizing fuzzy system with crisp system to construct the next generation computer system.
  • 梅谷 陽二, 吉田 和哉
    1990 年 26 巻 2 号 p. 188-195
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    This paper investigates operational performance of space manipulators installed on a free-flying robot satellite, by difining and analyzing its workspace and manipulability measure.
    The one-to-one correspondence is not found between joint space and inertial task space in space free-flying manipulators, therefore, their workspaces should be defined with some additional restrictions on a base satellite motion or an operation path of manipulator hand. According to these conditions, the authors classify and define five types of workspaces, which are new and important concept for operating space manipulators.
    A manipulability measure is defined by the generalized Jacobian matrix, which describes kinematic and dynamic property of a space free-flying manipulator. The manipulability of space manipulators is evaluated lower than that of ground-fixed manipulators, because it is reduced by dynamical coupling between the manipulator motion and the base satellite.
    Operation posture and configuration design problems for a two dimensional robot satellite model are discussed with respect to wide workspace and high manipulability.
  • 則次 俊郎, 和田 力, 西口 登
    1990 年 26 巻 2 号 p. 196-203
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In assembly, deburring, griding, not only the position and speed control but also the force control is important. A pneumatic servo seems to be effectively applied to this kind of constrained work, because its high compliance due to the air compressibility is useful to the force control.
    In the force control system, the characteristics of the object intervenes in the closed loop, so its influence has to be considered in the design of controller. However, it is generally unknown and often changes during the operation. To cope with these problems, more intelligent control schemes are required.
    In this study, to construct the force control system capable of accommodating to unknown objects and any other changes of operating conditions, the adaptive control strategy is applied. The controller is designed with the adaptive pole-placement method to avoid the non-minimum phase problem.
    Firstly, the influence of the object stiffness on the force control performance of the pneumatic servo system is analyzed. Secondly, the structure of the adaptive control system compensating for a disturbance is described. Thirdly, the adaptability to the unknown controlled plant, the effect of the compensation for disturbance, the robustness to changes of the object stiffness and the operating pressure and the force control accuracy are experimentally examined. As a result, it can be verified that the adaptive control scheme is very effective to improve the force control performance of a pneumatic servo. Lastly, matters to be attended to decide some control system parameters necessary to install the adaptive control system are considered.
  • 川村 貞夫, 宮田 慶一郎, 花房 秀郎, 石田 公成
    1990 年 26 巻 2 号 p. 204-210
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    We propose a new method to control pneumatic drive systems. In this control scheme, pressure feedback loops are effectively used in order to improve the performance of pneumatic systems. Concretely speaking, pressure feedback loops and position feedback loops are hierarchically connected, and the desired pressure patterns are composed of the position feedback input. The stability of the motion is investigated by using a Liapunov function. As the result, it can be mathematically proven that the motion of pneumatic drive systems converges to the desired point as time tends to infinity even if the disturbance affects the motion. The effectiveness of this control scheme is demonstrated through simulation results.
  • 押切 正浩, 出口 光一郎, 田村 安孝, 赤塚 孝雄
    1990 年 26 巻 2 号 p. 211-218
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    “2-Level Weighted Vector Quantization” is proposed as an efficient data compression method for images, without losing edge information. A conventional vector quantizer, such as the LBG-algorithm, does not incorporate a special procedure for retaining edge information which is very important for the under-standing of images. Therefore, the codebook for our quantizer is generated from weighted training vectors according to the contents of the edge information. To suppress degradation of the smooth part of the image, in our procedure the vector quantizer is used sequentially for both the locally averaged and the residual image. To evaluate the quality of the compressed image, we designed an index which is based on the number of edges and the sharpness of each edge in a local area. An experiment was performed to quantize the actual image and the reconstructed image was evaluated according to this index. The results showed that our method achieved a higher score than the LBG-algorithm or the adaptive discrete cosine transform.
  • 丁 全鋼, 湯浅 秀男, 伊藤 正美
    1990 年 26 巻 2 号 p. 219-224
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, an approach to plan collision-free motions for general robot manipulators based on the artificial potential function in joint space is proposed. The potential function is defined as the intersection of the obstacles and the path which is described by some sine functions in the joint space. The collision-free path is obtained by minimizing the value of this potential function using the steepest decsent method. Since the obtained path is constructed by the sums of some sine functions, it is ensured to be smooth. The proposed algorithm is applied to the movement of the manipulators with two and three links and its effectiveness is shown by computer simulation.
  • 平井 慎一, 浅田 春比古, 得丸 英勝
    1990 年 26 巻 2 号 p. 225-232
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    A model-based approach to the monitoring of assembly processes is presented. In assembly operations, efficient control strategies cannot be generated by single control laws because the task condition varies in the process. Robot manipulators must monitor the process state in order to modify task strategies depending on the state. In this paper, we develop a new method of generating procedures that identify the state of assembly processes. The procedure interprets displacement and force information acquired in the process in order to identify the process state. In particular, the system is capable of monitoring the contact state; at which surface or edge the assembly parts are contacting to each other.
    We first analyze kinematic and static behavior of assembly parts at each contact state by applying the theory of polyhedral convex cones. The theory enables us to treat constraints by mechanical contact in a systematic manner. Identification procedures that discriminate contact states are formulated with use of the polyhedral convex cones. We then develop a new method of generating the procedures automatically based on geometric models of assembly parts. To reduce real-time computations, the procedures are simplified to a minimum set by using reduction rules of polyhedral convex cones.
  • 米澤 洋, 黎 友盛, 西村 行雄
    1990 年 26 巻 2 号 p. 233-235
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    This paper proposes a new method of digital control system design based upon series expansion, and the newly weighted sum of square error betweenn reference model and control system is introduced. Numerical example is given to illustrate the effectiveness of the proposed method.
  • 藤井 信夫, 竹内 尚寿, 羅 正華
    1990 年 26 巻 2 号 p. 236-238
    発行日: 1990/02/28
    公開日: 2009/03/27
    ジャーナル フリー
    In this paper, a lumped parameter model for a visco-elastic system is proposed and examined through some experiments. The experimental results show that the lumped parameter model is a good approximation for the original distributed parameter system in certain situation.
feedback
Top