巻号一覧
39 巻, 9 号
選択された号の論文の11件中1～11を表示しています
- |<
- <
- 1
- >
- >|
-
山本 広樹, 青島 伸治2003 年 39 巻 9 号 p. 793-799
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーA concept manipulating multiple signals from three axes sensors flexibly is presented. To begin with, the expression form based on a quaternion is applied to the three channel measurement signals from a sensor in order to stand for those scalar signals with a single signal channel in real system. The imaginary part of the quaternion signal is filled with three signals from the sensor, and an empty signal is added as the real part at the front end of the processing system. Therefore, to convey one channel signal such system has four signal lines inside. Those signal lines are not equivalent sense. Secondary, a method to decompose a quaternion signal into a series of signals expressed by periodic quaternion functions and its inverse one are led. Those operations could be performed simply with the transform equation according to arithmetic operation rules on quaternion. Such transformation and its inverse one seem to be formal extension of complex Fourier transform of two dimensional system. Hence, when this transformation is introduced, it may be possible to design or consider some functional linear system by using the quaternion signal on frequency domain like as complex signal processing method. Lastly attending this concept to a FIR filter which is extended to quaternion signal, then the characteristics of this extended filter are analyzed. As the result, composite effect, for instance low pas filtering and the changing the distribution of signal power to each axes components, are observed. Also, it seems to be not difficult to find a proper quaternion parameters for getting the desired function on the system. From those matters, it may be thought that the quaternion signal processing will have various possibilities of useful applications.抄録全体を表示PDF形式でダウンロード (1998K) -
寺本 顕武, 鶴田 浩輔2003 年 39 巻 9 号 p. 800-807
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a defect characterization method by spatio-temporal gradient analysis over the Lamb wave field. The surface ultrasound waves that propagate in the direction of the layer are potential candidates of non-destructive evaluation (NDE) methodology. The proposed method has an ability to classify the surface acoustic wave field through the rank of the covariance matrix defined over the four-dimensional vector space which is spanned by following components: a vertical displacement, its vertical velocity, and a pair of shear strains of the surface. The covariance matrix provides the information about defects. The determinant of the covariance matrix, therefore, is proposed as the index of homogeneity of the object surface. In this study, the computational process in wave field near the defects is discussed and their physical meanings are investigated through FDTD-simulations and acoustic experiments.抄録全体を表示PDF形式でダウンロード (3499K) -
高見 弘, 辻野 太郎, 藤井 隆雄2003 年 39 巻 9 号 p. 808-816
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a quantitative evaluation for robust stability and robust performance of an ILQ optimal current-control system for permanent magnet synchronous motors via μ-analysis method. The μ-analysis is carred out by introducing a unique cost-function for robust performance and formulating a general servo-system which enables us to treat both ILQ and PI servo-systems evenly. By comparing with conventional PI control, it is quantitatively confirmed that the ILQ control is excellent in both robust stability and robust performance for variations of the motor's parameters. The results obtained give analytical verifications for simulation results which we had already reported.抄録全体を表示PDF形式でダウンロード (2191K) -
Ying SHEN, Kazuhiko TERASHIMA, Ken'ichi YANO2003 年 39 巻 9 号 p. 817-826
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents a model based on the load position of a rotary crane, validated by experimental results. Specifically, it provides a method for controlling the sway of a rotary crane using a straight transfer transformation (STT) model. The STT model was built and its parameters were geometrically derived. Taking into account variations in rope length, the optimal control of a rotary crane for reduction of residual sway was solved using the Davidon-Fletcher-Powell (DFP) optimization method. The proposed control method using the STT model was demonstrated to be effective in eliminating the influence of centrifugal force through simulation and experiments.抄録全体を表示PDF形式でダウンロード (1964K) -
川邊 武俊, 城 新一郎, メンスレ ミシェル2003 年 39 巻 9 号 p. 827-834
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーFor less energy consumption, toroidal continuously variable transmission (TCVT) is used for automobile power train systems. In order to compensate non-linearity and time variant properties of TCVT, gain scheduled nonlinar servo controller has been designed, and its effectiveness has been verified through computer simulation results and test results using actual TCVT.抄録全体を表示PDF形式でダウンロード (1912K) -
忻 欣, 兼田 雅弘2003 年 39 巻 9 号 p. 835-840
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーThis paper studies the energy based control for swinging up two inverted pendulums on a cart. This paper presents an original and detail analysis of the convergence of the energies of two pendulums and provides the condition such that the pendulums can be swung up. Moreover, the motion of two pendulums under the energy based control is investigated and the stability analysis of the motion is carried out via the concept of stability of an invariant set. All these analyses not only show theoretically the effectiveness of the energy based control for swinging up two pendulums, but also pave an way for analyzing and designing the energy based control for more complicated underactuated mechanical systems.抄録全体を表示PDF形式でダウンロード (1650K) -
村田 純一, 梶原 義龍, 平澤 宏太郎2003 年 39 巻 9 号 p. 841-847
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーA neural network is proposed based on a divide-and-conquer scheme. The network has gates which control firing of its hidden nodes. By opening and closing the gates depending on input values, the network divides the input space into sub-regions and assigns its nodes to each of them to produce the desired output in that region. The division mechanism is constructed by learning. A new learning method is proposed which divides the space in accordance to the difficulties; areas with larger errors are divided into smaller sub-regions. Thus, the nodes in the network are more densely assigned to areas with higher difficulties to ‘conquer’ the areas appropriately. Function approximation examples are provided to illustrate the validity of the proposed network.抄録全体を表示PDF形式でダウンロード (1924K) -
永井 康史, 土屋 茂, 木村 真一2003 年 39 巻 9 号 p. 848-856
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーWe applied auditory information as well as visual information to achieve safe and reliable operation of the robot arm aboard Engineering Test Satellite VII (ETS-VII). In our earlier experiments, we found that performance of operators depends on their skill levels and on their respective degrees of familiarity with the user interface which they were usually using by the operator.
Previous studies of man-machine interfaces for use with space robots have mainly been focused applying visual information to on improve usability. We examined the relation between the skill levels of operators and the user interface. We conducted experiments with the user interface in a variety of modalities with and without auditory information to supplement the visual information. We used an eye-mark recorder (EMR) to assess the performance of the operators, because it is said that eye movement provides a good representation of the human mental state. EMR was used to measure the eye movements of human operators during the experiments. The result was that fixation times and spectra of the velocity frequency of eye movements in the observation of status display which provide information on the space robot differed according to the skill levels of operators.抄録全体を表示PDF形式でダウンロード (4395K) -
近藤 敏之, 伊藤 宏司2003 年 39 巻 9 号 p. 857-864
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーIn recent robotics fields, much attention has been focused on utilizing reinforcement learning for designing robot controllers, since environments the robots will be situated could be unpredictable for human designers in advance. However there exist some difficulties, one of them is well known as “curse of dimensionality problem” in which as a state space for a learning system (e.g. robot) becomes continuous and high dimensional, the learning process results in time-consuming. Therefore, so as to adopt reinforcement learning for such complicated systems, not only adaptability but also computational efficiencies should be taken into account. In this paper, an evolutionary state recruitment strategy for an actor-critic reinforcement learning system based on NGnet is proposed, which enables the learning system to divide/rearrange its state space gradually according to the task complexity and the progress of learning. Simulation results and real robot implementations of a peg pushing robot control task show the validity of the method.抄録全体を表示PDF形式でダウンロード (2785K) -
Yuedong XU, Nobuo SANNOMIYA, Yajie TIAN2003 年 39 巻 9 号 p. 865-871
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーThe shifting bottleneck is well known as a method for solving complex scheduling problems. Based on the theory of constraints, the efficiency of the production system depends on the bottleneck process. The shifting bottleneck is to optimize the bottleneck process one by one and consequently the efficiency of the total system is improved. In this paper a new shifting bottleneck method is proposed for solving flexible flow shop problems. In this method the original problem is decomposed into several parallel machine scheduling problems to be solved one by one. The disjunctive graph is used to specify the schedule, and the tabu search is used to solve the parallel machine scheduling problems. A computation experiment is performed for two kinds of problems which are the problems generated randomly and the problems derived from a real production line. The effectiveness of the proposed method is shown in the computation result.抄録全体を表示PDF形式でダウンロード (1132K) -
川崎 晴久, 上木 諭, 伊藤 聡2003 年 39 巻 9 号 p. 872-878
発行日: 2003/09/30
公開日: 2009/03/27
ジャーナル フリーThis paper presents an adaptive coordinated control method for multiple robot arms grasping a common object. In the proposed controller, the dynamic parameters of both object and robot arms are estimated adaptively. The desired motions of the robot arms are generated by an estimated object reference model. The controller needs measurements of only positions and velocities of the object and robot arms and doesn't need the measurements of forces and moments at contact points. The asymptotic stability of motion is proven by the Lyapunov-Like Lemma. Experiment of the two robot arms handling a common object is shown.抄録全体を表示PDF形式でダウンロード (2467K)
- |<
- <
- 1
- >
- >|