• Title/Summary/Keyword: Keio University

Search Result 211, Processing Time 0.028 seconds

Composite Neural Networks for Controlling Semi-Linear Dynamical Systrms: Example from Inverted Pendulum Problem

  • Yamamoto, Yoshinobu;Anzai, Yuichiro
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 1989.10a
    • /
    • pp.1129-1134
    • /
    • 1989
  • In this paper, we propose a neural network for learning to control semi-linear dynamical systems. The network is a composite system of four three-layer backpropagation subnetworks, and is able to control inverted pendulums better than systems based on modern control theory at least in some ranges of parameters. Three of the four subnetworks in our network system process angles, velocities, and positions of a moving inverted pendulum, respectively. The outputs from those three subnetworks are input to the remaining subnetwork that makes control decisions. Each of the four subnetworks learns connection weights independently by backpropagation algorithms. Teaching signals are given by the human operator. Also, input signals are generated by the human operator, but they are converted by preprocessors to actual input data for the three subnetworks except for the network for control decisions. The whole system is implemented on both of 16 bit personal computers and 32 bit workstations. First, we briefly provide the research background and the inverted pendulum problem itself, followed by the description of our composite neural network model. Next, some results from the simulation are given, which are subsequently compared with the results from a control system based on modern control theory. Then, some discussions and conclusion follow.

  • PDF

Approximate Analysis of MAC Protocol with Multiple Self-tokens in a Slotted Ring

  • Sakuta, Makoto;Sasase, Iwao
    • Journal of Communications and Networks
    • /
    • v.5 no.3
    • /
    • pp.249-257
    • /
    • 2003
  • Ring networks are very commonly exploited among local area and metropolitan area networks (LAN/MAN), whereas cells or small fixed-size packets are widely used in synchronized ring networks. In this paper, we present an analytical method for evaluating the delay-throughput performance of a MAC protocol with multiple self-tokens in a slotted ring network under uniform traffic. In our analysis, we introduce the stationary probability, which indicates the number of packets in a node. Also, it is assumed that each node has a sufficiently large amount of self-tokens, and a slotted ring has the symmetry. The analytical results with respect to delay-throughput performance have similar values to computer simulated ones. Furthermore, in order to achieve fair access under non-uniform traffic, we propose an adaptive MAC protocol, where the number of self-tokens in a node dynamically varies, based on the number of packets transmitted within a specified period. In the proposed protocol, when the number of packets transmitted by a node within a specified period is larger than a specified threshold, the node decreases the number of self-tokens in a per-node distributed method. That results in creating free slots in the ring, thus all nodes can obtain an equal opportunity to transmit into the ring. Performance results obtained by computer simulation show that our proposed protocol can maintain throughput fairness under non-uniform traffic.

Empirical Comparison of Word Similarity Measures Based on Co-Occurrence, Context, and a Vector Space Model

  • Kadowaki, Natsuki;Kishida, Kazuaki
    • Journal of Information Science Theory and Practice
    • /
    • v.8 no.2
    • /
    • pp.6-17
    • /
    • 2020
  • Word similarity is often measured to enhance system performance in the information retrieval field and other related areas. This paper reports on an experimental comparison of values for word similarity measures that were computed based on 50 intentionally selected words from a Reuters corpus. There were three targets, including (1) co-occurrence-based similarity measures (for which a co-occurrence frequency is counted as the number of documents or sentences), (2) context-based distributional similarity measures obtained from a latent Dirichlet allocation (LDA), nonnegative matrix factorization (NMF), and Word2Vec algorithm, and (3) similarity measures computed from the tf-idf weights of each word according to a vector space model (VSM). Here, a Pearson correlation coefficient for a pair of VSM-based similarity measures and co-occurrence-based similarity measures according to the number of documents was highest. Group-average agglomerative hierarchical clustering was also applied to similarity matrices computed by individual measures. An evaluation of the cluster sets according to an answer set revealed that VSM- and LDA-based similarity measures performed best.

Fundamental Study on Effect of Preceding Vehicle Information on Fuel Consumption Reduction of a Vehicle Group

  • Matsumoto, Shuichi;Kawashima, Hironao
    • Journal of Communications and Networks
    • /
    • v.15 no.2
    • /
    • pp.173-178
    • /
    • 2013
  • It is a concern that eco-driving vehicles, because their driving behavior differs from other vehicles due to e.g. e-start, may inhibit smooth traffic flow. Therefore, it is necessary to study the cooperative eco-driving done by a vehicle group, putting "vehicle-to-vehicle communication" and "road-to-vehicle communication" into perspective. Based on these factors, this study aimed to: 1) Analyze fuel consumption rates and driving behaviors of more than one vehicle following an Eco-Driving vehicle. 2) Examine the effect of information on the fuel consumption rate of the preceding vehicles on the following vehicles. As a result, the following findings were obtained: 1) By providing information to multiple following vehicles, the fuel consumption rate of the second vehicle was not lowered, while that of the third one was. 2) It is possible that, when information on fuel consumption of a preceding vehicle is provided to the following one, an inter-vehicular distance is shortened during deceleration to contribute to smooth traffic flow. From the above results, it is suggested that, when targeting a vehicle group, sharing the information on preceding vehicles is effective.

Preparation of Textured Bi0.5(Na,K)0.5TiO3-BiFeO3 Solid Solutions by Reactive-Templated Grain Growth Process

  • Kato, Kyoko;Kimura, Toshio
    • Journal of the Korean Ceramic Society
    • /
    • v.43 no.11 s.294
    • /
    • pp.693-699
    • /
    • 2006
  • Textured $Bi_{0.5}(Na,K)_{0.5}TiO_3-BiFeO_3$ ceramics were prepared by the reactive-templated grain growth process, using platelike $Bi_4Ti_3O_{12}$ particles. The effects of chemical composition in $Bi_{0.5}(Na,K)_{0.5}TiO_3$ on texture development and densification were examined. Textured ceramics were obtained by using $Bi_{0.5}K_{0.5}TiO_3$ as an end member of the solid solution but densification was limited. Dense ceramics were obtained by using $Bi_{0.5}Na_{0.5}TiO_3$ but texture did not develop. Dense, textured ceramics were obtained by using $Bi_{0.5}(Na_{0.5}K_{0.5})_{0.5}TiO_3$.

IPv6 Autoconfiguration for Hierarchical MANETs with Efficient Leader Election Algorithm

  • Bouk, Safdar Hussain;Sasase, Iwao
    • Journal of Communications and Networks
    • /
    • v.11 no.3
    • /
    • pp.248-260
    • /
    • 2009
  • To connect a mobile ad hoc network (MANET) with an IP network and to carryout communication, ad hoc network node needs to be configured with unique IP adress. Dynamic host configuration protocol (DHCP) server autoconfigure nodes in wired networks. However, this cannot be applied to ad hoc network without introducing some changes in auto configuration mechanism, due to intrinsic properties (i.e., multi-hop, dynamic, and distributed nature) of the network. In this paper, we propose a scalable autoconfiguration scheme for MANETs with hierarchical topology consisting of leader and member nodes, by considering the global Internet connectivity with minimum overhead. In our proposed scheme, a joining node selects one of the pre-configured nodes for its duplicate address detection (DAD) operation. We reduce overhead and make our scheme scalable by eliminating the broadcast of DAD messages in the network. We also propose the group leader election algorithm, which takes into account the resources, density, and position information of a node to select a new leader. Our simulation results show that our proposed scheme is effective to reduce the overhead and is scalable. Also, it is shown that the proposed scheme provides an efficient method to heal the network after partitioning and merging by enhancing the role of bordering nodes in the group.

Three-Dimensional Numerical Analysis for Detonation Propagating in Circular Tube

  • Sugiyama, Yuta;Matsuo, Akiko
    • Proceedings of the Korean Society of Propulsion Engineers Conference
    • /
    • 2008.03a
    • /
    • pp.364-370
    • /
    • 2008
  • Spinning detonations propagating in a circular tube were numerically investigated with a one-step irreversible reaction model governed by Arrhenius kinetics. The time evolution of the simulation results was utilized to reveal the propagation mechanism of single-headed spinning detonation. The track angle of soot record on the tube wall was numerically reproduced with various levels of activation energy, and the simulated unique angle was the same as that of the previous reports. The maximum pressure histories of the shock front on the tube wall showed stable and unstable pitch modes for the lower and higher activation energies, respectively. The shock front shapes and the pressure profiles on the tube wall clarified the mechanisms of two modes. The maximum pressure history in the stable pitch remained nearly constant, and the single Mach leg existing on the shock front rotated at a constant speed. The high and low frequency pressure oscillations appeared in the unstable pitch due to the generation and decay of complex Mach interaction on the shock front shape. The high frequency oscillation was self-induced because the intensity of the transverse wave was changed during propagation in one cycle. The high frequency behavior was not always the same for each cycle, and therefore the low frequency oscillation was also induced in the pressure history.

  • PDF

Dual-innervated multivector muscle transfer using two superficial subslips of the serratus anterior muscle for long-standing facial paralysis

  • Sakuma, Hisashi;Tanaka, Ichiro;Yazawa, Masaki;Oh, Anna
    • Archives of Plastic Surgery
    • /
    • v.48 no.3
    • /
    • pp.282-286
    • /
    • 2021
  • Recent reports have described several cases of double muscle transfers to restore natural, symmetrical smiles in patients with long-standing facial paralysis. However, these complex procedures sometimes result in cheek bulkiness owing to the double muscle transfer. We present the case of a 67-year-old woman with long-standing facial paralysis, who underwent two-stage facial reanimation using two superficial subslips of the serratus anterior muscle innervated by the masseteric and contralateral facial nerves via a sural nerve graft. Each muscle subslip was transferred to the upper lip and oral commissures, which were oriented in different directions. Furthermore, a horizontal fascia lata graft was added at the lower lip to prevent deformities such as lower lip elongation and deviation. Voluntary contraction was noted at roughly 4 months, and a spontaneous smile without biting was noted 8 months postoperatively. At 18 months after surgery, the patient demonstrated a spontaneous symmetrical smile with adequate excursion of the lower lip, upper lip, and oral commissure, without cheek bulkiness. Dual-innervated muscle transfer using two multivector superficial subslips of the serratus anterior muscle may be a good option for long-standing facial paralysis, as it can achieve a symmetrical smile that can be performed voluntarily and spontaneously.

A study on fuzzy control of manipulator with artificial rubber muscles (고무인공근 매니퓰레이터의 퍼지제어에 관한 연구)

  • ;Keio Watanabw;Nakamura, Masatoshi
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 1993.10a
    • /
    • pp.1047-1051
    • /
    • 1993
  • A fuzzy controller of a manipulator with artificial rubber muscles is proposed. The fuzzy logic controller as a compensator is described to control the trajectory tracking of a -two link manipulator, where computed torque control method has already assumed to be applied. We shows that the fuzzy compensator with a simple adaptive scaling technique is effective for the robust control when there exist model uncertainties and/or untuned feedback gains. The effectiveness of the proposed control method is illustrated by some experimental results for a circular path tracking.

  • PDF

On the mater plan of the green in Kamakura City

  • Koshimizu, hajime;Ishikawa, Mikik;Koshizawa, Akira;Tsuchiya, Shiro
    • Journal of the Korean Institute of Landscape Architecture International Edition
    • /
    • no.1
    • /
    • pp.84-88
    • /
    • 2001
  • Decision progress, plan content and planning division title of the master plan of the green in Kamakura City Japan were described. Conservation of the green and various effectiveness of the measure for realizing the green creation plan and the future development were discussed. It was indicated that the cooperation between citizens land owners and administration of the country, prefecture and city was important.

  • PDF