By Topic

Systems, Man, and Cybernetics, 1997. Computational Cybernetics and Simulation., 1997 IEEE International Conference on

Date 12-15 Oct. 1997

Go

Filter Results

Displaying Results 1 - 25 of 92
  • 1997 IEEE International Conference On Systems, Man, And Cybernetics

    Page(s): 0_3 - 0_46
    Save to Project icon | Request Permissions | PDF file iconPDF (4066 KB)  
    Freely Available from IEEE
  • Author index

    Page(s): 0_47 - 0_60
    Save to Project icon | Request Permissions | PDF file iconPDF (1083 KB)  
    Freely Available from IEEE
  • Power command enhancement in mobile communication systems using an embedded fuzzy unit

    Page(s): 4364 - 4369 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (594 KB)  

    We propose a fuzzy logic-based power command enhancement scheme for the mobile station of a cellular phone system. By defining the necessary linguistic variables of the power commands and corresponding fuzzy inference rules, an embedded fuzzy unit (EFU) is set up. The role of our EFU is to produce appropriate digital power commands instead of one-bit discrete commands directly obtained from the base station. Simulations show that the EFU-based mobile station can generate considerably smoother received power at the base station. The overshoot and channel tracking error are also small. In addition, our method has the advantage of simplicity, and is thus easy to be realized by hardware. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Combining neural network, genetic algorithm and symbolic learning approach to discover knowledge from databases

    Page(s): 4388 - 4393 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (500 KB)  

    Classification, which involves finding rules that partition a given data set into disjoint groups, is one class of data mining problems. Approaches proposed so far for mining classification rules for databases are mainly decision tree based on symbolic learning methods. In this paper, we combine artificial neural network, genetic algorithm and symbol learning methods to find classification rules. Some experiments have demonstrated that our method generates rules of better performance than the decision tree approach and the number of extracted rules is fewer than that of C4.5 View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A topological map based navigation system for mobile robotics

    Page(s): 4394 - 4399 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (544 KB)  

    We discuss two forms of cognitive map which are commonly used in mobile robotics when navigation requirements and environmental conditions make them appropriate; namely global metric, and topological. Whilst the former is a more powerful representational form which is suitable for building route plans in static environments, the latter is typically much less computationally intense and may be embedded directly into a behavior-based platform which can cope with local dynamics efficiently. We then focus on a method for extracting a topological map from a metric one, ie. a transformation from spatial non-procedural to symbolic procedural knowledge. Results from simulation experiments are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A concurrency characteristic in Petri net unfolding

    Page(s): 4266 - 4273 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (608 KB)  

    Unfolding, originally introduced by McMillan (1995), is gaining ground as a partial-order based method for the verification of concurrent systems without state-space explosion. However, it can be exposed to redundancy, which may increase its size exponentially. So far, there have been trials to reduce such redundancy resulting from conflicts by improving McMillan's cut-off criterion. In this paper, we show that concurrency is also another cause of redundancy in unfolding, and we present an algorithm to reduce such redundancy in live, bounded and reversible Petri nets, which is independent of any cut-off algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using a distance metric on genetic programs to understand genetic operators

    Page(s): 4092 - 4097 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (492 KB)  

    I describe a distance metric called “edit” distance which quantifies the syntactic difference between two genetic programs. In the context of one specific problem, the 6 bit multiplexor, I use the metric to analyze the amount of new material introduced by different crossover operators, the difference among the best individuals of a population and the difference among the best individuals and the rest of the population. The relationships between these data and run performance are imprecise but they are sufficiently interesting to encourage further investigation into the use of edit distance View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intelligent hybrid multi-agent computational architecture for complex systems

    Page(s): 4471 - 4476 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (412 KB)  

    Intelligent Multi-Agent Hybrid Distributed Architecture (IMAHDA) for complex systems has been defined at the task structure level and the computational level. IHMADA has been built around five information processing phases. The five phases represent a deliberate as well as automated or reflexive reasoning structure. This paper outlines some aspects of the computational level of IMAHDA. The computational level is derived from four layers, namely, object layer, software agent layer, intelligent agent layer, and problem solving agent layer. Agent description in terms of percepts, goals, actions, and environment for some of the problem solving agents is also outlined. IMAHDA has been applied in engineering (real time systems) and management areas View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reversibility verification of Petri nets using unfoldings

    Page(s): 4274 - 4278 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (320 KB)  

    Discrete event systems are modeled compactly by using Petri nets (T. Murata, 1989). Although Petri nets have powerful mathematical verification ability, in many cases, we have to construct the whole state space for verification. A reachability graph is one of the representations of the state space. Petri nets can describe huge systems compactly, but sometimes it is impossible to generate the reachability graph because of an explosion in required computational time and space. A Petri net is called a reversible net, when it can come back to the initial marking from any reachable marking. The paper considers a verification method for reversibility. An unfolding is obtained by unfolding a Petri net, and it preserves reachability information of an original net and structural analysis on it is much easier than on the original net. The article clarifies relations between unfoldings and reversibility, and provides a verification method for reversibility by using unfoldings View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The use of cultural algorithms with evolutionary programming to control the data mining of large-scale spatio-temporal databases

    Page(s): 4098 - 4103 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (544 KB)  

    We use an evolutionary computational approach based upon cultural algorithms to guide the incremental learning decision trees by ITI. The results are compared to those produced by ITI itself for a complex real-world database. The results suggest that ITI can indeed produce optimal trees in some cases, and can produce optimal trees using an evolutionary approach in others View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robust control of uncertain measure differential systems with delay: Riccati-like approach

    Page(s): 4332 - 4336 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (236 KB)  

    In this paper, we present a novel design approach of robust controller for a class of measure differential systems with delay. To design the controller and analyze the stability of closed-loop dynamics, the Riccati-like approach is employed and a novel transformation of state is introduced. In our approach, it is unnecessary to assume the norm bound of constant matrix of state variable x(t-r) to be small. To illustrate the effectiveness of the scheme, a numerical example is given finally View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The adaptive weight using RAM

    Page(s): 4053 - 4056 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (312 KB)  

    This article analyses the saturation problem of a RAM neural network, a n-tuple classifier containing 340 12-input neurons applied to the character recognition task, using the British mail data bank. It presents data to evaluate this problem and correlates it to other characteristics of the RAM nets. Therefore, two novel approaches were suggested to reduce the network saturation and improve the recognition level: the filtered RAM and the adaptive weight using RAM (AWURAM). The first version simply multiplies each input vector by a digital filter during the training and the recall phases. The second approach associates the weight concept to the network in order to distinguish different regions among the trained classes View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Competitive co-evolution based game-strategy acquisition with packaging strategies method

    Page(s): 4418 - 4423 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB)  

    One of the important issues in the field of artificial intelligence is to develop the method on the acquisition of game strategy. In this paper, our purpose is the acquisition of game-strategy by using competitive co-evolution approach as a search method. The competitive co-evolution is the mechanism of the interactive improvement. However, in practical game, it isn't easy to acquire the best strategy by applying competitive co-evolution model, because of complex strategy space. Therefore, to design the acquisition system of stronger game strategy, we propose an improved competitive co-evolution model that introduces the concept of “package” as a set of good strategies. Creating the good package needs to collect some good strategies to defeat various kinds of the opponents. To show the efficiency of this algorithm, we apply it to a complicated game View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • KOLEM: studies on evolution of expertise in a society of learning agents

    Page(s): 4465 - 4470 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (356 KB)  

    A society of agents is one in which many agents coexist and interact with each other while carrying out their tasks. Each agent builds a certain level of expertise. Thus the interactions among agents would help each agent to carry out its task better by tapping into the collective knowledge assets of the society. Agents may differ in the ways in which they interact with other agents. As the society evolves, agents may change from one type to another. This kind of an environment presents interesting dynamics. We describe a simulator that models such a society of agents. Using a package delivery domain, agent interactions, their relative performances and the evolution and performance of the society as a whole are studied View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Facial analysis applying an attentive processing strategy

    Page(s): 4069 - 4074 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (860 KB)  

    A technical realization of an attentional mechanism localizing and analyzing prominent facial regions in gray-level images is presented. By adopting the gaze control principles of the HVS for developing an image processing system, a complex image analysis problem may be decomposed into a number of subproblems which can be solved step by step in a simpler way. The attentive processing strategy starts with the localization of the prominent facial regions based on a saliency representation carrying all the information needed to select and restrict the extent of the prominent facial regions. Subsequently, the detected and now spatially limited regions are classified to evaluate the benefit of applying more detailed analysis methods. During the detailed analysis step the exact positions of anatomical landmarks or keypoints such as eye and mouth centers are determined. Fundamental to the attentive processing strategy is that all processing modules consider and process only prominent and really characteristic image structures. Only those images structures are considered which contribute to the solution of the actual processing task. The attentive processing strategy proposed is able to cope with variations due to the perspective angle or pose, orientation, illumination, and contrast of the studied facial images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • What is a symbolic measurement process?

    Page(s): 4139 - 4145 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (612 KB)  

    Numbers have emerged historically as the most popular/convenient form of representation, and the basic scientific paradigm is built on the foundation of numeric, or quantitative, concepts. Measurement, as conventionally understood, is the corresponding process for (numeric) representation of objects or events. Any (including a future) measurement device is constructed based on the underlying mathematical structure that is thought appropriate for the purpose. It has gradually become clear that the classical numeric mathematical structures, and hence the corresponding (including all present) measurement devices, impose on “real” events/objects a very rigid form of representation, which cannot be modified dynamically in order to capture their combinative, or compositional, structure. To remove this fundamental limitation, a new mathematical structure-evolving transformation system (ETS)-was proposed earlier. This mathematical structure allows one to capture inductively the compositional structure of events. In other words, the model defines a new form of object representation-struct-that captures the combinative, or symbolic, structure of events. Moreover, this combinative structure, which represents what appears to be the only possible natural symbiosis of the discrete/symbolic and the continuous representations, in principle, cannot he captured by the classical mathematical structures. The authors briefly discuss the concept of a fundamentally new, biologically inspired, measurement process, inductive measurement process, based on the ETS View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • User modelling and information filtering for consumer health information

    Page(s): 4207 - 4210 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB)  

    This paper describes an information filtering system that creates personalized web pages with new and relevant information for users. The system maintains a model of each user based on web pages visited and relevance rankings made of those pages made by the user. The system has been used with a consumer health information web site. Focus interviews with test subjects were conducted. Among the most important issues raised by test subjects was a sense of loss of control and privacy. Human factors issues and future implications of intelligent agents in the consumer health care sector are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Applying morphological filters to acoustic broadband correlograms

    Page(s): 4182 - 4187 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (624 KB)  

    An acoustic broadband correlogram, a two-dimensional display over time of the cross-correlation of signals received by a two-sensor system, is often used to detect targets close to the sensor system. Often, the correlogram is cluttered by the strong interference generated by distant and yet strong sources. These strong interferers appear in the form of dark vertical and nearly-vertical lines in the broadband correlogram and hinder the detection of nearby targets. This paper presents an automatic target detection system for broadband acoustic correlograms designed to take into account the interference from strong sources. The system consists of three parts: (1) a pre-processing stage, (2) a Delay Curve Hough Transform (DCHT) and (3) an onion peeling process. The pre-processing stage consists of a morphological filter (the rolling arc) and a linear histogram stretching process to improve the contrast of the morphologically filtered correlogram. The modified Hough Transform, based on the assumption that the target moves with constant velocity, is designed to detect delay curves in the broadband correlograms. Finally, the onion peeling process is an iterative process that sequentially removes the strongest detected delay curve from the correlogram and feeds the modified correlogram back to the Hough Transform to detect the next strongest delay curve View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Weight value testing in artificial neural networks

    Page(s): 4023 - 4028 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (540 KB)  

    This paper presents a method for testing ANN interconnection weight values in programmable ANNs. The methodology requires separate weights for excitatory and inhibitory contributions to a neuron's activation. In the test the excitatory and inhibitory weights are set to be equal, producing no net contribution to the activation. Any faults in the weight values result in an imbalance that is amplified by the neuron activation function and is easily detectable at the neuron output. Analysis is performed on the detection approach that includes fault detection thresholds as a function of neuron gain, network size, and weight perturbation. A methodology is outlined that uses the proposed testing approach followed by example test results View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A framework for the study of interactions between robots

    Page(s): 4297 - 4302 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (748 KB)  

    The paper describes a conceptual and experimental framework which provides a general centralized/decentralized control architecture for the study of visual interactions between multiple mobile robots. Two examples of situations using this framework are presented: the first one studies interactions between robots in pursuit/evasion problems. The second one concerns decentralized, vision based, control strategies for handling a collaborative manipulation View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimization of Earth station receiving performances to minimize total system cost

    Page(s): 4370 - 4375 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (380 KB)  

    SCPC and TDMA satellite communication systems of meshed network configurations are studied. In SCPC, Earth station transmission power of a channel is controlled to minimize transponder power, keeping the required communication quality. With TDMA, transmission burst bit rate is controlled to maximize transponder capacity. Using cost models of Earth stations and transponders, sets of equations to optimize receiving performances (G/T) of Earth stations are derived. The equations are solved and the total system cost including a transponder and Earth stations is minimized View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Considerations on extended feature vectors in automatic face recognition

    Page(s): 4075 - 4080 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (600 KB)  

    Clearly, automatic recognition in large face populations will require many measurements. There have been few approaches which aim to generate such extended feature vectors. One approach considered combining several different sets, including feature descriptions and transform components, with apparent advantages accrued by the orthogonality of the measurements. More robust measures have included a new technique for eye location which employs concentricity using only few parameters and requiring little a priori information concerning a face's location. Further, a dual contour employing global energy minimisation, again requires few parameters to provide measurements describing the face's boundary, again aimed at inclusion within an extended feature vector. Naturally, we seek to capitalise on minimal statistical correlation to improve recognition capability. To this end, we consider further the analysis of potential advantages of orthogonality, and show how this can indeed improve recognition capability. Accordingly, there is much research potential in extending the feature vector for automatic face recognition: there are rich avenues for future research in generation and combination of feature vectors for use in large face populations View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Supporting situation awareness in aviation systems

    Page(s): 4177 - 4181 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (472 KB)  

    Situation awareness is recognized as one of the most critical aspects in the aviation domain. Many features of our high technology environment can act to subtly degrade situation awareness, however, including high levels of complexity, out-of-the-loop performance decrements resulting from automation, and lack of synergy in human and machine decision making. An approach that seeks to design for situation awareness enhancement is explored as a viable alternative to traditional automation approaches View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • EIKONES-an object-oriented language for image processing

    Page(s): 4188 - 4191 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (436 KB)  

    This paper presents an object oriented language, called EIKONES, for efficient and flexible image processing. The EIKONES language provides the user flexibility and friendliness for image processing which are not available in other image processing tools. The basic idea behind EIKONES is the consideration of the image processing algorithms as objects and the appropriate development of a formal grammar for its actual implementation. Results are provided by using the EIKONES language View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Comparison of projection-based face recognition algorithms

    Page(s): 4057 - 4062 vol.5
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (504 KB)  

    Traditional methods of evaluating face recognition algorithms report on their performance on a single set of images. However, this does not fully capture algorithm performance or allow for statistical comparisons between algorithms. To overcome this problem, we present an evaluation methodology that allows one to obtain a robust and comprehensive assessment of face recognition algorithms. From the evaluation methodology, we formulate a model that allows for statistical hypothesis testing. We use this to compare the performance of different algorithms of the same test set, and one algorithm on different test sets View full abstract»

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