By Topic

Systems, Man and Cybernetics, IEEE Transactions on

Issue 3 • Date Mar 1995

Filter Results

Displaying Results 1 - 14 of 14
  • Development of distributed problem solving systems for dynamic environments

    Page(s): 400 - 414
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1596 KB)  

    As knowledge-based techniques are introduced in supervision and control applications, the need for a structured, methodological approach in the development of these systems increases. Current knowledge-based systems development methodologies are founded on individualistic models of expertise, and cannot face the challenge of interacting with humans or other systems in dynamic application environments. This article introduces a development method that combines models of expertise with interaction-based distributed artificial intelligence. The applicability of the method is demonstrated on an example of supervision of a power transmission network. The analysis of the proposed system is based on the KADS methodology. The design phase is structured in a series of steps: first, the domain description is used to select a model of distributed problem solving. Then, problem constraints and a set of design rules guide the mapping of the analysis' entities to a design solution. Design decisions are justified at every step. Alternative design proposals are experimentally compared with the aid of a multi-agent platform and a simulator. The experimental results indicate possible design refinements. Finally, the generality of the approach is discussed and future research directions are indicated View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The assignment heuristic for crossing reduction

    Page(s): 515 - 521
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (668 KB)  

    Several applications use algorithms for drawing k-layered networks and, in particular, 2-layered networks (i.e. bipartite graphs). Bipartite graphs are commonly drawn in the plane so that all vertices lie on two parallel vertical lines, and an important requirement in drawing such graphs is to minimize edge crossings. Such a problem is NP-complete even when the position of the vertices on one layer is held fixed. This paper presents a heuristic, called the assignment heuristic, for edge crossing minimization in bipartite graphs, which works by reducing the problem to an assignment problem. The main idea of the assignment heuristic is to position simultaneously all the vertices of one layer, so that the mutual interaction of the position of all the vertices can be taken into account. We also show that the idea underlying the assignment heuristic can be effectively applied in other cases requiring edge crossing minimization View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new approach to synthesizing free motions of robotic manipulators based on a concept of unit motions

    Page(s): 453 - 463
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (948 KB)  

    A concept of unit motions and a new scheme for synthesizing free motions of robotic manipulators are proposed. Unit motions are defined as smooth primitive motions in the hand coordinate system by employing normalized uniform B-spline functions. Complex motions are then generated as a superposition of time-shifted and weighted unit motions. Corresponding to unit motions, a concept of unit control inputs is also introduced and an entire control input is then obtained in exactly the same way as complex motions are generated from unit motions. Moreover, in order to build further complex motions from simpler ones in a systematic way, operations on motions are defined including spatial, temporal and structural ones. Based on unit motions and unit control inputs, we then develop a formal method for motion planning and control for robotic manipulators. Motions are represented formally as the time series of weight vectors, or equivalently as the so-called control polygons, and the operations on motions are then regarded as operations on control polygons. Simulation examples are included to demonstrate how to generate motions from unit motions and operations, and properties of the resulting motions are also described View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new methodology for designing a fuzzy logic controller

    Page(s): 505 - 512
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (600 KB)  

    A new methodology is proposed for designing a fuzzy logic controller (FLC). A phase plane is used to bridge the gap between the time-response and rule base. The rule base can be easily built using the general dynamics of the process, and then readily updated to contain the delayed information for reducing the deadtime effects of the process. An adaptive gain method is also proposed to help the database design and the controller tuning. Much of the FLC design can be shifted to the design and tuning of gain. A good performance can be achieved both in transient state and steady state without use of multidecision tables. Application of FLC with these new methodologies is presented for a thermal process with a varying deadtime to show the robust performance of FLC and the effectiveness of these methodologies View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fast neural learning and control of discrete-time nonlinear systems

    Page(s): 478 - 488
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (840 KB)  

    The problem of learning control for a general class of discrete-time nonlinear systems is addressed in this paper using multilayered neural networks (MNNs) with feedforward connections. A suitable extension of the concept of input-output linearization of discrete-time nonlinear systems is used to develop the control schemes for both output tracking and model reference control purposes. The ability of MNNs to model arbitrary nonlinear functions is incorporated to approximate the unknown nonlinear input-output relationship and its inverse using a new weight learning algorithm. In order to overcome the difficulties associated with simultaneous online identification and control in neural networks based adaptive control systems, the new learning control architectures are developed for both adaptive tracking and adaptive model reference control systems with online identification and control ability. The potentials of the proposed methods are demonstrated by simulation examples View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Integrating field work in system design: a methodology and two case studies

    Page(s): 385 - 399
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1284 KB)  

    Bridging the “gap” between observation of end-users' needs and support system design is a prerequisite for any user-centered design and support system development process. The authors describe a framework, functional information and knowledge acquisition (FIKA-) modeling, that can be used both to structure observations in complex environments and to structure support system functionality. The central idea is to focus on information and knowledge types an end-user needs rather than on specific end-user actions. Two examples from industrial process control illustrate the approach. The first example demonstrates how high-level system requirements can be captured. The second example shows how to gather knowledge to generate a model specifying end-users' information and knowledge requirements View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Operator scheduling approaches in group technology cells-information request analysis

    Page(s): 438 - 452
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1212 KB)  

    This paper examines the group technology (GT) scheduling problems, in which each cell is staffed with one operator. The role of the operator is to load jobs into machines, to unload jobs from machines, and to transport jobs between machines. Since the productivity of the cell is directly proportional to the level of productive work performed by the operator, the manner in which the operator moves between machines affects productivity. Although the optimal operator schedule/walking-pattern has been shown to be the best operator schedule for well-structured production environments, the underlying assumptions of the optimal schedule often limit its practical implications. With the current popularity of gust-in-time production, the problem of finding an efficient operator schedule/walking-pattern has become a critical issue in shop floor scheduling. According to the information requirements, this paper devises three heuristic approaches: a scheduling rule, multiple scheduling rules (MSR), and cycle switching rules (CSR). This paper shows that there is a tradeoff between the quality of the resulting schedule and the information contents of heuristics used View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Inference of a probabilistic finite state machine from its output

    Page(s): 424 - 437
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1200 KB)  

    This paper addresses the problem of identifying a probabilistic FSM from its output. The authors propose a formal description of the FSM and its output, using a regular phrase-structured grammar. This description is associated with a cost measuring its information content. The authors then state the inference problem as a combinatorial optimization problem with an objective function define as the cost of the description of the FSM and its output. A heuristic algorithm is proposed which processes the output “on line” and yields a local minimum of the authors' criterion. At each step (i.e. as new data is observed and processed), the procedure searches locally through the space of non-probabilistic FSMs, i.e., the transitions are initially regarded as being only present or absent. When a non-probabilistic model has been generated, the transition probabilities are determined from their relative frequencies given the behavior indicated by the data. This yields maximum likelihood estimates of the probabilities. The costs of the obtained probabilistic FSMs are then computed, and the K minimum cost FSMs are kept as starting points for the next step search. At each step, the generation of the FSMs is done via a local search through the neighborhoods of the K best FSMs obtained at the previous step. This algorithm is compared with similar algorithms proposed previously, and tested on a range of examples. It is found to work for a wider range of FSMs than the previous methods, and it is much more practical for large problems than previously proposed exhaustive search techniques View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On testing of sequential machines using circuit decomposition and stochastic modeling

    Page(s): 489 - 504
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1220 KB)  

    Test generation for sequential circuits has been a difficult task. This is due to the large search space to be considered in test pattern generation. In this paper the detection of permanent faults in sequential circuits by random testing is analyzed utilizing the circuit partitioning approach together with a continuous parameter Markov model. Given a large sequential circuit, it is partitioned into several smaller partitions using either series or parallel decomposition. For each partition with certain stuck faults specified, the original state table and its error version are derived from an analysis of the partition under fault-free and faulty conditions, respectively. A random testing strategy that uses a three-state Markov model is used for detecting permanent stuck faults. Experimentation on various sequential circuits has shown that a significant saving in testing or test generation time can be achieved if we partition the circuit and then test each of its components as opposed to testing the circuit in its original form View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A high speed systolic architecture for labeling connected components in an image

    Page(s): 415 - 423
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (820 KB)  

    Connected components detection and labeling is an essential step in many image analysis techniques. The efficiency of the connected components labeling algorithm is critical for many image processing and machine vision applications that require real time response. The advances in the areas of parallel processing and VLSI technology can be exploited in designing hardware algorithms of high speed and throughput. In this paper, the authors propose a systolic algorithm and architecture for finding connected components in an image. The architecture is simple and can be implemented as a special purpose VLSI chip. Although, the algorithm has a time complexity of O(N2), this is in terms of the actual clock cycle which is estimated as 25 nano seconds. The proposed hardware can process a 128×128 image in 0.85 msec and uses 128 processors whereas the MPP requires 94.6 msec with 16384 processors. The only special purpose hardware that exists requires 300 msec to label a 512×512 image which can be accomplished in 13.5 msec using the authors' proposed hardware View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Iterative histogram modification of gray images

    Page(s): 521 - 523
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (408 KB)  

    We develop an iterated histogram modification algorithm for enhancing gray images. The modified image has very few gray levels corresponding to the spikes found. A process of sharpening the peaks on the histogram of an image is considered using an iterative process where large bins increase at the expense of nearby smaller bins View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A sensor-based navigation for a mobile robot using fuzzy logic and reinforcement learning

    Page(s): 464 - 477
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1268 KB)  

    The proposed navigator consists of an avoidance behavior and goal-seeking behavior. Two behaviors are independently designed at the design stage and then combined them by a behavior selector at the running stage. A behavior selector using a bistable switching function chooses a behavior at each action step so that the mobile robot can go for the goal position without colliding with obstacles. Fuzzy logic maps the input fuzzy sets representing the mobile robot's state space determined by sensor readings to the output fuzzy sets representing the mobile robot's action space. Fuzzy rule bases are built through the reinforcement learning which requires simple evaluation data rather than thousands of input-output training data. Since the fuzzy rules for each behavior are learned through a reinforcement learning method, the fuzzy rule bases can be easily constructed for more complex environments. In order to find the mobile robot's present state, ultrasonic sensors mounted at the mobile robot are used. The effectiveness of the proposed method is verified by a series of simulations View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Accelerated template matching using template trees grown by condensation

    Page(s): 523 - 528
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (760 KB)  

    Template trees provide a means of accelerating nearest neighbor searches for problems in which KD-trees and similar data structures do not work well because of the high dimension and/or sophisticated distance function used. Suppose the points for which nearest neighbors are being sought are noisy 16×24 images of characters. Each point has 16×24=384 dimensions. Images which a good distance function would classify as similar may have very different values at a dozen or more randomly chosen pixels. Template trees work directly with the distance function rather than with the 384 components of the points. An algorithm is presented for selecting templates from a set of training points, and organizing them into a template tree which is guaranteed to correctly identify all of the training points. The tree construction algorithm is similar in many ways to the condensation algorithm for template selection, although it organizes templates into a tree as it selects them. A tree containing approximately 2000 images of capital letters was constructed using a training set of about 8000 points. Using the tree, an average of only about 140 point to point distance calculations were needed to identify an unknown image. Identification accuracy was comparable to that obtained using 2000 templates without a tree View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptive training algorithm for back-propagation neural networks

    Page(s): 512 - 514
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (308 KB)  

    A dynamic learning rate for back-propagation training of artificial neural networks is proposed as a weighted average of direction cosines of the incremental weight vectors of the current and previous steps. Experiments on training an EEG-based sleep state pattern recognition scheme have demonstrated its improved performance View full abstract»

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