巻号一覧
21 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
- |<
- <
- 1
- >
- >|
-
大山 恭弘, 古田 勝久1985 年 21 巻 2 号 p. 109-115
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーIn spite that an RC helicopter has various advantages due to its small size and wireless operation, because of its inherent instability, some automatic flight control system is desirable to be implemented in order to fly it desirably. In this paper, a control system for the RC helicopter with five degrees of freedom which can be used for the training simulator is proposed and implemented by minicomputer. The design technique based on discrete Model Following Servo control system, where the criteion function is specified in continuous time is proposed. This technique is effective when the sampling time can't be decided a priori or has to be changed due to the change of program because the designed closed loop characteristic is equivalent to the continuous one. Reference models whose time scale is changed and whose degrees of freedom are restricted are used. The simulator can mimic the motion of RC helicopter and the validity of the model and the control system is confirmed experimentally.抄録全体を表示PDF形式でダウンロード (1291K) -
Mehdi NOURI-SHIRAZI, Nobuo SANNOMIYA, Yoshikazu NISHIKAWA1985 年 21 巻 2 号 p. 116-122
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーWe consider a discrete linear dynamical system with known Gaussian system disturbances and known non-Gaussian observation noises. We propose an approximation filter which is optimal in the sense that the trace of the estimation error-covariance matrix is minimum for the optimal score function belonging to a preassigned class of score functions. Then, a recursive system of equations is derived for a scalar dynamical system which gives a lower and an upper bound of the filter's performance. In order to illustrate the properties of the proposed filter, a scalar system is considered. Results of simulation are presented for two cases; a) mixture Gaussian observation noise, and b) Cauchy observation noise. The lower and upper bounds of the filter's performance are also included for the case b.抄録全体を表示PDF形式でダウンロード (824K) -
山根 裕造1985 年 21 巻 2 号 p. 123-130
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーAs mentioned before, a class of system observer can afford to provide simulteneously the state estimates as well as the parameter estimates of an linear unknown system. In this paper, an adaptive control input can be implemented so that the plant output estimate might become identical to the output of the reference model. As a result, the following severel advantages can be shown.
[1] Simpler configuration of overall closed loop than several previous procedures.
[2] Minimum convergence time of the plant output to the output of the reference model.
[3] Not necessarily reqirement of identifying uniquely actual parameters of unknown plant in the process of actuating the adaptive control system combined with the system observer.
[4] Judgement whether the adaptive system works properly
[5] A numerical computational work for a 29-th order plant as a demostration of a higher order plant is done to illustrate the basic idea.抄録全体を表示PDF形式でダウンロード (854K) -
内門 茂, 金井 喜美雄, 杉山 博之1985 年 21 巻 2 号 p. 131-136
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper deals with the problem of an adaptive control system for linear discrete-time multivariable systems having unknown disturbances.
The features of this method are described as: the resulted system doesn't construct the servo system explicitly and by introducing the equality equation which makes the term of disturbances unobservable apparently, we can eliminate the effect of disturbances. Therefore, the proposed designing method has a very simple procedure and we can use the existing adaptive control scheme having no disturbance. Moreover, the signals in the adaptive loop remain bounded and the output errors tend to zero asymptotically.
Simulation studies are shown to justify the proposed scheme.抄録全体を表示PDF形式でダウンロード (660K) -
知能自動車の人工眼広瀬 武志1985 年 21 巻 2 号 p. 137-144
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper reports on an artificial eye of intelligent vehicle. The author had presented a method to recognize obstacles in the path of the vehicle and materialized it as a road pattern recognition system. The system employed a stereo TV camera assembly and a hard wired logic unit. It was capable of recognizing obstacles in the recognition region which was approximately 6.4m wide, 1m deep and 9m ahead the vehicle.
However, the depth of the region was not enough to make the vehicle run smooth. Then, this time, the author made an attempt to expand the recognition region. The improved method makes use of a fact that the recognizable area changes its location when either of the signals from 2 sets of TV camera is delayed. The TV signals are delayed in different 32 ways and all the delayed signals are processed in parallel to detect presence of obstacles in continuous 32 recognizable areas. The whole recognition region covers an area of about 16m wide, 20m deep and 5m ahead the vehicle. It is divided into sections, 128 horizontally and 16 deeply. Presence of obstacles on each section is checked, and 2 dimensional distribution of obstacles is figured out. Then, the clear path for the vehicle is computed from the 2 dimensional distribution of obstacles, and output to a problem solving device. The total time to process a vision is 36ms.
The road pattern recognition system was equipped on the intelligent vehicle and put to experiments on a test course. The system functioned successfully to recognize 2 dimensional distribution of obstacles such like other vehicles and roadside guardrails.抄録全体を表示PDF形式でダウンロード (2661K) -
佐藤 勝俊, 内田 勉, 畑中 浩1985 年 21 巻 2 号 p. 145-150
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper is primarily directed toward analyzing the single pulse response in pneumatic lines.
The exact solution to the single pulse response was obtained by using the complicated transfer function derived by Iberall, and compared with the approximate one which was derived from the Brown's approximate transfer function. The discrepancy with the exact and the approximate is small, so the approximate solution is adopted to compare the experimental result with the theoretical one.
In this experiment, the input signal was generated by the single pulse generator with the disc chopper. In order to compare the theoretical wave forms with the experimental ones, the traces obtained from the oscilloscope were photographed. The diameter, the length of lines, and input pressure amplitude were varied to investigate their effects. As the input signal wave form was not rectangular but rather trapezoidal, the analysis for trapezoidal input was also presented. Experimental result shows the fairly good agreement with the theory based on the trapezoidal input.
As a result, it is found that the Brown's approximate transfer function is very useful in analyzing the transient response of pneumatic lines. Furthermore, in this experiment, the effect of the input pressure amplitude (3.2, 4.8, 6.4kPa) was not observed.抄録全体を表示PDF形式でダウンロード (697K) -
流量変動が大きい場合の誤差と流量波形による影響金住 順二, 杉田 尚弘1985 年 21 巻 2 号 p. 151-156
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーIn a previous paper, the author discussed the problem of “pulsation error” of the float-type flowmeter, and gave an approximate analytical expression for the error assuming smallness of the pulsation. In the present study the same problem is pursued for large amplitude case of pulsation. The equation of motion of the float derived previously is solved by making use of a numerical method. A comparison of the resulting solution with the previous one shows that the formula for the pulsation error obtained in the previous paper can practically be applicable up to the case of fairly large pulsation. The effect of wave form (such as chopping wave, square wave and sawtooth wave) to the pulsation error is also examined.抄録全体を表示PDF形式でダウンロード (739K) -
小林 史典, 近久 直一, 小林 彬1985 年 21 巻 2 号 p. 157-163
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーThough the measurement of minute fluctuations in rotational speed is involved in various fields of engineering, eccentricity error unavoidable in practical situations poses limits on accuracy. This paper proposes a method for providing high accuracy while easing the setup procedure.
The principal configuration of the instrument is that the image of a radial stripe pattern attached to the rotating shaft of interest is projected by a lens onto the polar-coordinate-type spatial filter. Undesirable fluctuation caused by the eccentricity of the pattern is canceled by the differential operation afforded by a number of elements of the spatial filter.
The primary advantage of the proposed method is its high accuracy in the order of 0.02% of the average, while employing relatively crude components. Also featured is the sensor housing separated by about 20cm from the rotating pattern, thus providing safety at high speeds.
Two examples of tortional vibration are evaluated: cogging of a motor, which could not be directly evaluated so far; and a planetary gear mechanism rotating at speeds as high as 15, 000rpm, which could not be traditionally treated because of the mechanical limitation of bearings.抄録全体を表示PDF形式でダウンロード (1955K) -
高橋 哲也, 藤村 貞夫, 豊田 弘道1985 年 21 巻 2 号 p. 164-171
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーMost of the pixel-by-pixel supervised classification methods for remotely sensed multispectral images are quite inefficient, because the discrimination procedure is applied to every pixel even in a uniform area consisting of the data belonging to only one class. As they make no use of the spatial properties of objects, they do not achieve high accuracy, either. As for the “per field” classification method, there is no established technique for segmentation of an image into “fields”.
In this paper is proposed a classification method where the spatial resolution for classification is adaptively varied according to the local uniformity of the image. The adaptive division of the area to be classified is realized on the assumption that the image has hierarchical (pyramid) structure. It is achieved by a two-step procedure, each step of which makes use of a statistical test of local uniformity of the image.
The classification accuracy and the processing time by this method are compared with those by the pixel-by-pixel maximum likelihood method which is most often used for classification of remotely sensed images. The comparison shows that this method improves the performance of classification both in efficiency and in accuracy.抄録全体を表示PDF形式でダウンロード (3416K) -
流動状内容物の運動シミュレーション梅谷 陽二, 伊能 教夫1985 年 21 巻 2 号 p. 172-176
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーThe peristalsis, commonly observed in animals' small intestines, serves skillful transport of the intraluminal contents. It is supposed that a neural control mechanism for peristalsis is immanent in the intestines, but the structure has not yet been made clear. Some researches on the peristaltic movement in view of biomechanics have been reported. However, they have given little attention to the fluid property of the actual intraluminal contents, such as chyme.
The authors carried out a study of the neural mechanism using a mechanical simulator which can handle actual fluid contents. The simulator was composed of a thin plastic tube corresponding to intestinal walls and 16 actuators spaced same short intervals on the tube. Each actuator contracts vertically and deforms the tube. The simulator was controlled by a microcomputer which performs neurodynamical simulation in real time while sensing fluid contents in the tube by mechanical sensors of the simulator.
Neural mechanism for peristalsis was researched by performing simulation experiments. And finally, a neural network model was obtained which can transport various viscid fluids. The simulation also showed that this model has a function of modifying the movement according to fluidity of contents.抄録全体を表示PDF形式でダウンロード (1395K) -
下条 誠, 石川 正俊1985 年 21 巻 2 号 p. 177-182
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーIn this paper the sensor which has a simple structure, flexibility and also has the function of displaying the pressure distribution is proposed, and measured results are shown.
The sensor consists of four layers. First and third layers are patterned electrodes. Second layer is the pressure conductive rubber whose resistance varies with pressure. Last layer is the liquid crystal whose color varies with temperature.
The theory of measurement is as follows: the distributed pressure(p) on the sensor causes a drop in the resistance (rp) of the pressure conductive rubber: as a voltage is applied between two patterned electrode layers. the electric current(i) passes through the rubber in proportion to the rp: then this electric current produces heat(q) and the temperature(θ) rises as the result. Consequently this changes the color of liquid crystal at the position. In a word, this sensor converts the intensity of pressure into the difference of temperature and displays it using liquid crystal. The following equations show these relations: θ∝q∝i2rp=v2/rp∝v2p∝p(ifrp∝p-1).
However, the heat diffusion occuring in the rubber and the electrodes causes the measurement errors. Therefore the ways of preventing heat from diffusion are necessary. Hence these methods are proposed; (1) the electrodes are constructed of a lot of small square electrodes which are connected with narrow channels each other. In order to decide the sizes of the patterned electrode, the simulation is carried out. (2) the heat produced in the rubber is controlled by using a heat controller.
This sensor is useful for the measuring the contact pressure between the soft materials, such as between human body and a seat.抄録全体を表示PDF形式でダウンロード (1401K) -
構造解析と制御系設計中村 仁彦, 花房 秀郎1985 年 21 巻 2 号 p. 183-189
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーIn this paper, we discuss three dimensional autonomous trajectory control (3D-ATC) of robot maniaulators based on sensory feedback. The 3D-ATC is effective to automate, for examples, welding, painting and measuring unknown shapes. However, the complete 3D-ATC has not yet been made practicable, because multivariable proximity sensors which measure full information between an endeffector and a desired trajectory have not been developed so far and the structure of the 3D-ATC system has not been analyzed sufficiently.
The 3D-ATC system is formulated as a control system in which only the relative position at present moment between an end-effector and a desired trajectory on an object is available for feedback information. And a controller for the 3D-ATC system is synthesized. Simulation is carried out to evaluate its effectiveness.抄録全体を表示PDF形式でダウンロード (988K) -
内山 勝, 清水 邦敏, 箱守 京次郎1985 年 21 巻 2 号 p. 190-196
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーIn order to design and control a robotic arm on the scientific basis, it is necessary to evaluate its manipulation performance. In this paper, the use of the Jacobian as the performance index is discussed from the kinematic and static standpoint of view.
The value of the Jacobian is related to the homogeneity of the transformation from the joint coordinates to the workspace coordinates and, therefore, can be an index for the manipulation skill at a point in the workspace. The index takes the minimum value at singular points at which the arm loses capability of moving in a certain direction and is least dexterous.
Color graphics display of the distribution of the index value gives an overall understanding of the manipulation performance. This is illustrated by numerical examples for two different types of robotic arms with six revolute joints.
As examples of the application of the index, trajectory planning and optimal design of arm mechanisms are described. With the index, it is possible to plan trajectories scientifically. Numerical results for avoiding singular points and turning a crank with the best hand orientation are presented. The average of the index value over the whole workspace can be an index for the comparison of different arm mechanisms. A two link arm is designed using the index.抄録全体を表示PDF形式でダウンロード (1450K) -
古川 敦, 菅沼 義昇, 伊藤 正美1985 年 21 巻 2 号 p. 197-204
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーThis paper presents a new methd to control biped locomotion robot. In general the motion of the constrained dynamic system, that corresponds to the double support phase in biped locomoyion, is described by using lagrange multiplier. In this paper we describe the system by using penalty function and propose a new method to control it. This method can equally control both the constrained system in the double support phase and the unconstrained system in the single support phase. and can control optionally the reaction force from the floor. Therfore it is useful to control not only biped locomotion but also manipulators.抄録全体を表示PDF形式でダウンロード (803K) -
趙 群飛, 戸苅 吉孝, 後藤 秀雄1985 年 21 巻 2 号 p. 205-207
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (256K) -
藤村 貞夫, 花泉 弘, 椿 広計1985 年 21 巻 2 号 p. 208-210
発行日: 1985/02/28
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (257K) -
1985 年 21 巻 2 号 p. 212
発行日: 1985年
公開日: 2009/03/27
ジャーナル フリーPDF形式でダウンロード (148K)
- |<
- <
- 1
- >
- >|