By Topic

Knowledge-Based Intelligent Electronic Systems, 1998. Proceedings KES '98. 1998 Second International Conference on

Date 21-23 April 1998

Go

Filter Results

Displaying Results 1 - 25 of 89
  • Author index

    Page(s): 617 - 619
    Save to Project icon | Request Permissions | PDF file iconPDF (163 KB)  
    Freely Available from IEEE
  • Optimization of computer-generated holograms by an artificial neural network

    Page(s): 220 - 223 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (284 KB)  

    Several computer-generated hologram (CGH) methods, such as the direct binary search, simulated annealing and genetic algorithm, have been proposed or used in order to decrease the quantum noise and reconstruction noise or to optimize the CGH. Since these methods are iterative approaches, they require long computation time to generate a CGH. In this paper, we propose a new method based on an artificial neural network (ANN) to reduce the high computation cost. In this scheme, we first use a couple of known optimized CGHs, which may be obtained by the traditional optimization methods, as teaching signals to train the ANN. With the trained ANN, we can easily and quickly obtain an optimized CGH without the optimization process for other input images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive-intelligent control by neural-network and genetic-algorithm systems and its application

    Page(s): 230 - 239 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (888 KB)  

    As a attempt towards brain-computing, adaptive-intelligent controls by neural-network and genetic-algorithm systems are studied. As an application the computer-simulation system for nano-structured materials is constructed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamic behaviors of an integrated circuit for recurrent neural networks

    Page(s): 260 - 267 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (748 KB)  

    In order to investigate dynamic behaviors of recurrent neural networks or asymmetric interconnection networks on neuro-chips, we design a hardware neural network with programmable synaptic weights according to the design rule of a CMOS technology. The full connections between neurons and the self-coupling can be performed. Some types of connections can produce many limit cycles on the network. The number of limit cycles increases sharply with increasing the number of neurons in case of nearest neighbor connections. As an example, there are at least 1.14×107 limit cycles in the case of 40 neurons. The limit cycles have basins of attraction, and hence, we may utilize the network as associative memory to retrieve dynamical cyclic patterns. After the SPICE simulation for the network, we fabricate the integrated circuit. The chip size is 4 mm×4 mm or 2.2 mm×2.2 mm. The main part of the chip has 49 synapses and 98 SRAM cells each two of which belongs to each synapse to store its weight. We present a procedure to construct the synaptic weights to produce particular limit cycles in a network. The procedure to make up a connection matrix is useful for hardware implementation in terms of the simple synaptic weights and its accuracy View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multimodal information retrieval, extraction and generation for use in the health domain

    Page(s): 307 - 316 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1432 KB)  

    If an organisation such as the Department of Defence is to maximise its use of information, it has to exchange and coordinate meanings between different kinds of information sources with information from a variety of domains. A multimodal meaning base is being developed that makes it possible to identify meanings that can be expressed in linguistic and cartographic modalities. The research involves the collection of multimodal resources including structured information (databases), unstructured information, the description of the semantics, and the expression in a computational resource. The outcome of the research is the development of a multimodal information management system, HINTS, that will help solve the problems of information exchange and coordination. Our domain is communicable diseases. The design of the multimodal information management tool utilised participatory design techniques involving researchers, designers, users and stake holders. Participatory design is a comprehensive and integrated methodology for designing graphical user interfaces using low-tech materials, object oriented methods, and frequent iterative usability testing. We extended this technique to include the design of the entire system. The techniques were effective across a multidisciplinary team that included researchers and clients View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • LPT scheduling on fuzzy tasks with triangular membership functions

    Page(s): 109 - 115 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (400 KB)  

    We (1998) demonstrated how discrete fuzzy concepts can easily be used in the LPT (longest processing time first) algorithm for managing uncertain scheduling. This paper extends application to the continuous fuzzy domain. The membership functions of the scheduled tasks are assumed to be triangular, and three heuristic fuzzy LPT scheduling methods are proposed to yield scheduling results with polygonal or triangular final completion-time membership functions. Experiments with different numbers of scheduled tasks are also presented to show the effectiveness of the proposed methods View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Interactive animation for visualisation of mapping of parallel programs into parallel processors

    Page(s): 386 - 395 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (684 KB)  

    Techniques of cartoon animation, when applied to the interface, provide the user with greater visual cues, and convey a feeling of substance to the objects being manipulated. The paper describes the effects used to animate many of the interactions in a structured graphical editor. The editor is used as a tool in a visual parallel programming environment and provides a new large application to test the effectiveness of the concept of animated direct manipulation. Also described are the implementation issues in order to provide animated graphical feedback for direct manipulation View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of temporal difference learning method to computer gaming

    Page(s): 36 - 39 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (244 KB)  

    Applies the temporal difference learning method and a neural network to the computer simulation game which deals with old Chinese history. Several computer simulation results show that the temporal difference learning method and neural network have been quite effective in order to let a player on the computer side have enough intelligence to cope with a strong human player View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Right brain computing hardware: a psychological brain model on silicon

    Page(s): 429 - 435 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (712 KB)  

    Despite the enormous power of present-day computers, digital systems cannot respond to real-world events in real time. Biological systems built with very slow chemical transistors, however, are very fast in such tasks as seeing, recognizing, and taking actions. The paper will discuss what we can learn from biological computing systems and how we can implement the principle using the state-of-the-art silicon technology. Intelligent LSI systems based on the psychological model of a brain are proposed. The system stores the past experience in nonvolatile vast analog memories and recalls the most similar event to a current input using the association processor working in the analog/digital-merged decision-making principle View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IC leadframe inspection system using multiple thread on the Windows NT

    Page(s): 282 - 285 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (312 KB)  

    In this paper we propose an automatic IC leadframe inspection system to inspect major defects in lead spacing, pad shift, tape position and punching defect. This system is composed of an image acquisition part and an inspection part for improvement of inspection speed and flexible inspection environment. We adopted Windows NT operating system in consideration of stability and compatibility of the system. The system has these 2 separate parts so that the image may be refreshed continually during inspection; hence, the whole system's performance was improved. A subpixel method using a high-resolution camera is adopted because the spacing inspection requires very high accuracy and this made it possible stably to maintain an accuracy of about one third of a pixel. Tape inspection is performed by using the mutual relation of the positions of the tape boundary lines. We developed an algorithm which can shrink and expand any polygon efficiently to inspect tapes of arbitrary shapes. By experiment using above inspection method, our inspection system was confirmed as high speed and high accuracy inspection system that can inspect about five leadframe units per second View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stochastic fuzzy servo control using multiple linear dynamic models

    Page(s): 474 - 482 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (564 KB)  

    A fuzzy servo system is described for a system with noises by using a stochastic fuzzy control method with some linear dynamic models. The fuzzy servo approach is applied to the control of rotational angle for an omnidirectional mobile robot with three orthogonal-wheel assemblies. The stochastic fuzzy servo method and its modified method with a static evaluation on the model confidence are implemented for some simulations of the mobile robot. In particular, two improvement methods are shown for the case when a set of the prespecified linear models does not include a model comparable to the reference rotational angle View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Sub-symbolic knowledge extraction environment for teaching process assistance

    Page(s): 411 - 417 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (772 KB)  

    We aim with this research is to build an integrated environment to serve as an assistant in the educational process. When we deal with unstructured knowledge, getting of useful information for the teaching process is very difficult. Neural networks can store subsymbolic knowledge, but until recently it was believed to be only in a “black-box” format. Knowledge extraction from NNs is a relatively new field, which tries to reduce these disadvantages and build a bridge between subsymbolic and symbolic knowledge. As the teaching process requires only symbolic knowledge, we believe this to be a chance for teachers to significantly improve their teaching materials and/or style by combining the symbolic knowledge of the domain theory with the rules extracted from the empirical subsymbolic knowledge stored in NNs trained on examples. Therefore, we developed a neural network's subsymbolic knowledge extraction environment for the teaching process assistance and also built a study case of teaching stock exchange developments View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Coordinative motion control for different kind of manipulators using a fuzzy compensator

    Page(s): 462 - 467 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (396 KB)  

    We describe a coordinative motion control based on the master/slave method for different kinds of manipulators. The master side consists of a pantagraph-type two link manipulator driven by a torque command, whereas the slave side is an industrial manipulator driven by a position command with a force/torque sensor. It is difficult to maintain a desired contact force by only a master/slave method. To maintain a desired contact force, a compensative position command is generated by using fuzzy reasoning with the force sensor information. The effectiveness is verified by experiments View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hidden partitioning of a visual feedback-based neuro-controller

    Page(s): 53 - 60 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (804 KB)  

    Robotic controllers take advantage from neural network learning capabilities as long as the dimensionality of the problem is kept moderate. The paper explores the possibilities offered by the combination of several neural networks to design more complex modular controllers. This modularity is based on an internal partitioning of the problem. The partitioning must remain hidden, and should not affect the controller's interface or functioning, including during its adaptation phases. We introduce a bi-directional architecture to derive the learning rules of the modules. The neuro-controller is trained globally, based on the interactions of the system with its environment, as one would do for a single network. The approach is illustrated on a robotic reaching application. Several partitioning variants of the neuro-controller are discussed and compared View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Moderatism: new concept for self-organization of neural networks

    Page(s): 246 - 249 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (304 KB)  

    A self-organization mechanism, “moderatism” is proposed, by which the input-to-output relationship of neural networks is automatically formed so as to adapt to the environment, even if there are feedback loops inside and/or outside. By applying this organization mechanism, it is shown that the networks acquire the reflex to avoid the dangerous stimulus. Also the networks start self-oscillation by themselves to get some amount of stimulus when there is no external signal View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Decision-making in fuzzy environment: a survey from stochastic decision process

    Page(s): 542 - 546 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (328 KB)  

    Interpreting the stochastic systems in a fuzzy environment as given by Bellman and Zadeh (1970) in the frame of stochastic decision process, we present two approaches for the multi-stage stochastic decision process: one is an invariant embedding method, and the other is an a posteriori-conditional decision process View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Qualitative interpretation of process trends by using neural networks

    Page(s): 2 - 6 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (352 KB)  

    Qualitative interpretation is a process to convert numerical output of sensors into symbolic representation. This process is one of the most critical path to connect intelligent systems with real world. In this paper, qualitative interpretation is realized as pattern-based classification of time-series signal by using ART2 neural networks. As an example, automatic classification of flow patterns in a pneumatic conveyor is successfully demonstrated View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Knowledge-based Doppler blood-velocimeter system

    Page(s): 334 - 339 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (456 KB)  

    A scheme of a knowledge-based Doppler blood-velocimeter system using an artificial neural network is proposed. The scheme supports feature extraction from a Doppler ultrasonography periodogram, and comparison with statistical data to make therapeutic studies on that basis. A physical model of blood flow with salient features taken into account, spectral estimation using the time varying autoregressive moving average (ARMA) model, and an artificial neural network (ANN) scheme for the determination of the coefficients of the ARMA model has been presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hybrid GA/SA approach to blind deconvolution

    Page(s): 144 - 149 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (428 KB)  

    A hybrid GA/SA (genetic algorithm/simulated annealing) approach is proposed for the blind deconvolution problem of image restoration. The blind deconvolution problem is modeled as an optimization problem whose cost function is to be minimized by the proposed hybrid approach. The approach combines the advantage of GA for global searches and the advantage of SA for local ones. The results indicate that it is possible to arrive at high quality solutions in reasonable time even for large-scale problems, such as image processing View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reasoning with tolerance

    Page(s): 418 - 427 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (892 KB)  

    This paper briefly discusses the existing MetaCon architecture, and its current upgrade to the ATTITUDE architecture for multi-agent, multilevel data fusion, as a motivation for formulating a tolerant approach to both procedural and declarative reasoning. Fuzzy first order logic is then formally defined before truth profiles are introduced to simultaneously accommodate ontological and epistemic uncertainty. Tolerant reasoning with truth profiles is then considered View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A more efficient man/machine interface: fusion of the interacting telethesis and smart wheelchair projects

    Page(s): 41 - 45 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (712 KB)  

    Presents a new and more efficient man/machine interface resulting from the fusion of two projects: the smart wheelchair project and the interacting telethesis project. This new interface allows a more natural control of the automatic motion capabilities of the smart wheelchair VAHM (autonomous vehicle for motor impaired people). Indeed, the pilot needs only to make his request (switch on the TV for example) without giving details about its location. The VAHM project aims to bring piloting assistance for motor impaired people to powered wheelchairs using techniques coming from mobile robotics. It is possible to call upon local primitives like “wall following” or “automatic forward motion”, or to plan an automatic trajectory. The second project aims to realize a man/machine communication interface for people with motor impairments; interface defined as an “interacting telethesis”, i.e. as a cognitive and/or valid motion resource(s) extension of the user. Therefore, the interacting telethesis allows disabled persons to be able to communicate (read, write, speak, listen, warn) with other people and to act on their environment (TV, music, etc.) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal stopping of a continuous-time dynamic fuzzy system under possibility theory

    Page(s): 553 - 558 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (344 KB)  

    An optimal stopping problem with fuzzy rewards is discussed in a continuous-time dynamic fuzzy system which is constructed under possibility theory. The fuzzy rewards are estimated by a fuzzy expectation derived from a fuzzy goal, and an optimal stopping time and the optimal fuzzy rewards are given View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Behavior-based search using small autonomous mobile robot vehicles

    Page(s): 294 - 301 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1256 KB)  

    This paper proposes a system that utilize a large number of autonomous intelligent “micro” systems to search and identify pre-defined objects. Robotic agents would be deployed in large numbers to collectively search an unknown area. The agents will search for objects of interest and upon locating the objects, either secure itself to that object and await further instruction from a base station or leave behind a marker indicating the object's location so that the object can be acquired later on. Simulation experiments on the deployment of these robotic agents have been conducted and the results of some of these experiments are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Experimental evaluation of an evolutionary scheduling scheme for the job shop scheduling problem

    Page(s): 197 - 201 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (436 KB)  

    Scheduling in a constrained environment is a complex and difficult process. Genetic algorithm (GA) approach has been increasingly used to solve various combinatorial optimization type problems. In this paper, we carry out experiments to investigate a GA based approach to the job shop scheduling problem (JSSP). We propose genetic heuristics to be used in our GA. We evaluate the performance of the proposed approach using well known benchmark instances for the JSSP View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of chaotic movements and fuzzy assessment of hands tremor in rehabilitation

    Page(s): 340 - 345 vol.3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (564 KB)  

    The analysis of chaotic components in movements during hands tremor and more classic analysis of the movements is considered in relation to its application to diagnosis and rehabilitation techniques. Fuzzy methods are used to characterize the movements and to assess their normality. Visual and auditory feedback is provided to help controlling the movements in a rehabilitation system that incorporates elements of virtual reality View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.