All issues
Volume 37, Issue 3
Displaying 1-12 of 12 articles from this issue
- |<
- <
- 1
- >
- >|
-
Masahiro SAEKI, Shogo TANAKA2001 Volume 37 Issue 3 Pages 185-192
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSOne approach to measure the temperature distribution in boilers is the measurement of the time-of-flight between acoustic sensors on the walls. The authors previously proposed a time-of-flight measurement system which made use of a pattern matching between actual and predicted sound waves, when sinusoidal pulse wave was transmitted from a speaker on one wall to an acoustic sensor on another wall. But, the method was a little troublesome, because it required to compensate for the predicted sound wave to realize a high accurate measurement of the time-of-flight.
This paper proposes a new approach measuring the time-of-flight by observing the phase difference between the transmitted and received signals (in stationary state) for several inputs of different frequencies. This yields the robustness on the existence of the noise. Simulations and experiments show the effectiveness of the proposed approach.View full abstractDownload PDF (2284K) -
Teturo ITAMI2001 Volume 37 Issue 3 Pages 193-202
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSFor the nonlinear Hamilton-Jacobi equation, we have proposed a new method to deal with the framework of linear theories. “Superposable complex waves” and a new constant HR repesenting the “strength” of the wave are introduced in this paper. Attendant to the optimal path we can assume a superposable complex wave in the state space subject to the “linear wave equation”. Within the weak limits of the wave the optimal path is obtained. This method is realized as an algorithm according to two factors; optimal control system as a constrained mechanical system, and its “linear operator” formalism using “Dirac-Brackets”. The real-part projection of the complex linear wave equation is shown to lead to a “generalized Hamilton-Jacobi equation”, where a term related to the wave is added to the Hamilton-Jacobi equation. The state feedback scheme is derived from this algorithm and a nonlinear system with 1-input 1-state variable is optimized by a simulation study using a typical method for problems of complex linear wave equation.View full abstractDownload PDF (2570K) -
Shuichi KUROGI, Takeshi NISHIDA2001 Volume 37 Issue 3 Pages 203-212
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSSo far, the controllers using the switching scheme of multiple models have been proposed for controlling the plants whose parameter values change occasionally and largely owing to wide varieties of environments involving different operating conditions, uncertainty, and so forth. Although the stability of the controllers has been analyzed, the efficiency of the models has not been examined in detail, where the present learning algorithm of competitive associative net called CAN2 is formalized and utilized to minimize the energy defined as the sum of identification errors of switched models so that the models can be efficiently utilized when the identification errors are large. In order for the CAN2 to learn and switch multiple models for reproducing the forward dynamics of the plant, two kinds of CAN2, called CAN2-1 and CAN2-2, are introduced. The former basically is for learning to achieve a piecewise linear approximation of nonlinear plant dynamics, and the latter basically is for learning to achieve adaptive linear approximation of time-varying plant dynamics. The CAN2-1 and CAN2-2 are embodied into the conventional model-based predictive controllers, which are called APC1 (adaptive predictive controller 1) and APC2, respectively. Computer simulation clarifies the behavior of the APC1 and APC2 and shows that the APC1 is useful for nonconvex nonlinear plants while the APC2 is for time-varying plants.View full abstractDownload PDF (2327K) -
Yasuhide KOBAYASHI, Toru ASAI2001 Volume 37 Issue 3 Pages 213-221
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper deals with synthesis problems for a class of delay systems, where the inputs of the delays can be measured and finite dimensional approximations for the delays are given with approximation errors. For the given problems, we propose a controller synthesis method based on gain scheduling for the approximation errors. The resultant controllers consist of finite dimensional systems with the delays. This structure is advantageous, since the resultant controllers can be implemented easily. In proposing the synthesis method, we show the following results: (1) There exists an equivalent problem where the output of the approximation error can be measured; (2) For the equivalent problem, there exists a controller which does not have the approximation error as scheduling parameter, even though the controller is synthesized via gain scheduled H∞ control for the approximation error.View full abstractDownload PDF (691K) -
Kei SENDA2001 Volume 37 Issue 3 Pages 222-228
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper proposes a dynamics-based control for redundant manipulators named the redundancy utilization in rate and acceleration (RURA) method. The RURA method calculates the desired joint trajectory satisfying the first subtask, i.e., a desired end-effector trajectory, and the required second subtask with an additional criteria to optimize the configuration. The second subtask is achieved by utilizing the redundancy in the motion rate and acceleration of joint variables. The RURA method then realizes a dynamics-based control based on a computed torque method. The RURA method can be extended to a resolved acceleration control and an impedance control. The RURA method is compared with the configuration control (CC) and the redundancy utilization in motion acceleration (RUMA) method that are commonly used. The RURA method overcomes problems of the CC and the RUMA method. Feasibility of the proposed method is examined by numerical simulations.View full abstractDownload PDF (1793K) -
Umerujan SAWUT, Tsuyoshi HANAMOTO, Teruo TSUJI2001 Volume 37 Issue 3 Pages 229-234
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSIn this paper, an identification and control method of a non-linear system with friction is proposed where a disturbance and sliding mode (SM) observer, and the genetic algorithm (GA) are used. The model of the nonlinear system used is a flexible arm system which is driven by a motor whose shaft has non negligible friction. The effectiveness of our proposal is confirmed by simulations and experimentsView full abstractDownload PDF (1768K) -
Masanori HARIYAMA, Fumitake YAMAGUCHI, Michitaka KAMEYAMA2001 Volume 37 Issue 3 Pages 235-241
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSFor high-speed collision detection, high-computational power is required in not only matching operation between pixel information of a robot manipulator and obstacles but also coordinate transformation. The proposed VLSI processor consists of content-addressable memories (CAMs) for parallel matching operation and processing elements (PEs) for parallel coordinate transformation. Design optimization can be attributed to minimization of the area-time product of a CAM and a PE under a condition of their 100% utilization. Since pixel information of each manipulator link is predetermined and not changed, a high-performance CAM based on a ROM cell is proposed for area-time-product reduction of a CAM. A PE based on a bit-serial pipelined architecture is also proposed for area-time-product reduction of a PE. A test chip is implemented in the 0.6μm CMOS double-metal technology. Its evaluation shows that the performance of the VLSI processor is three orders of magnitude higher than that of a general-purpose microprocessor.View full abstractDownload PDF (2737K) -
Takanori FUKAO, Ryosuke OMURA, Norihiko ADACHI2001 Volume 37 Issue 3 Pages 242-249
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSQ-learning is one of the famous algorithms for reinforcement learning. A usual way for expressing a Q-value function is using a Q-table which is a look-up table. But it is difficult to specify the discretize size of the state spaces without prior knowledge. In this paper, the method of the adaptive construction of state spaces on Q-learning by storing the data an agent has experienced is proposed. The effectiveness of this method is confirmed by some simulations of path-planning problems. Furthermore, the method of automatically setting the parameter for resolving the trade-off between exploration and exploitation is proposed.View full abstractDownload PDF (2074K) -
Masao SAKAI, Noriyasu HONMA, Kenichi ABE2001 Volume 37 Issue 3 Pages 250-254
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper demonstrates that the Lyapunov exponents of recurrent neural networks can be controlled by our proposed methods. One of the control methods minimizes a squared error eλ=(λ-λobj)2/2 by a gradient method, where λ is the largest Lyapunov exponent of the network and λobj is a desired exponent. λ implying the dynamical complexity is calculated by observing the state transition for a long-term period. This method is, however, computationally expensive for large-scale recurrent networks and the control is unstable for recurrent networks with chaotic dynamics since a gradient collection through time diverges due to the chaotic instability. We also propose an approximation method in order to reduce the computational cost and realize a “stable” control for chaotic networks. The new method is based on a stochastic relation which allows us to calculate the collection through time in a fashion without time evolution. Simulation results show that the approximation method can control the exponent for recurrent networks with chaotic dynamics under a restriction.View full abstractDownload PDF (1404K) -
Akihiko SUGIKI, Shoshiro HATAKEYAMA, Katsuhisa FURUTA2001 Volume 37 Issue 3 Pages 255-261
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSThis paper presents a relay feedback system that is equivalent to the Chua circuit. The circuit is referred to as being one of well-known chaotic models in the chaos field and its chaotic nature has been confirmed both theoretically and experimentally. It is shown in this paper that bilinear transformation allows describing the Chua circuit as the chaotic relay feedback system. Benefits associated with the relay feedback design include: (a) In the sampling and zero-order hold discretization, numerical simulation reveals that a double-scroll chaotic attractor is generated for larger sampling-periods compared to the original system, (b) The nonlinear feedback characteristics of Chua's system can be implemented using simple relay circuit.View full abstractDownload PDF (1868K) -
Shoji YOSHIKAWA, Ken'ichi NISHIGUCHI2001 Volume 37 Issue 3 Pages 262-267
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSNavigation, guidance and control (GN&C) based on imagery information is a key technology in lunar landings. A digital elevation map (DEM) is needed to verify GN&C algorithms. We propose a new method of generating a DEM that looks similar to one made from past measurements and at the same time allows a magnified use for lower altitudes. In the proposed method, small ups and downs, which are lost in a coarse DEM made from photographic measurements, are extrapolated by fractal modeling. Furthermore, small craters are added to match their statistical distributions from data on previous measurements. It is applied to a moon DEM. The obtained DEM shows that the small ups and downs and craters lost in the original DEM are restored; this restoration looks quite similar to the observation photos. Its georama is produced by laser lithography, and its photo also looks similar to the observation photos.View full abstractDownload PDF (2945K) -
Susumu HASHIZUME, Katsuaki ONOGI, Yoshiyuki NISHIMURA2001 Volume 37 Issue 3 Pages 268-270
Published: March 31, 2001
Released on J-STAGE: March 27, 2009
JOURNAL FREE ACCESSWe have formulated a condition/event net (C/E net) construction problem, discussed its solvability and showed the basic idea as to how it is solved. However, a practical solution technique has not been presented yet. In many manufacturing systems, there exists various periodic operations. In this paper, we deal with the C/E net construction problem with specification which exhibits periodic behavior. We show the existence of solutions of the problem by presenting a method of solving the problem.View full abstractDownload PDF (507K)
- |<
- <
- 1
- >
- >|