Search

The long-wave time evolution of technoeconomic development is used to model a theoretical cybernetic framework useful in understanding and predicting recurrent social phenomena. This analysis suggests the technoeconomic system is not chaotic but exhibits cyclical and periodically deployed information.

In this paper, an alternative training approach to the EEM-based training method is presented and a fuzzy reactive navigation architecture is described. The new training method is 270 times faster in learning speed; and is only 4% of the learning cost of the EEM method. It also has very reliable convergence of learning; very high number of learned rules (98.8%); and high adaptability. Using the rule base learned from the new method, the proposed fuzzy reactive navigator fuses the obstacle avoidance behavior and goal seeking behavior to determine its control actions, where adaptability is achieved with the aid of an environment evaluator. A comparison of this navigator using the rule bases obtained from the new training method and the EEM method, shows that the new navigator guarantees a solution and its solution is more acceptable. Index Terms - Behavior fusion, fuzzy logic, goal seeking, neural network, obstacle avoidance, reinforcement learning, vehicle navigation.

The idea of combining the network of HMM's and the dynamic programming-based search is highly relevant to online handwriting recognition. The word model of HMM network can be systematically constructed by concatenating letter and ligature HMM's while sharing common ones. Character recognition in such a network can be defined as the task of best aligning a given input sequence to the best path in the network. One distinguishing feature of the approach is that letter segmentation is obtained simultaneously with recognition but no extra-computation is required. Index Terms - Cursive script, hidden Markov model, ligature, network search, online character recognition, segmentation, Viterbi algorithm.

An algorithm for computing three-dimensional (3-D) velocity field and motion parameters from a range image sequence is presented. It is based on a new integral 3-D rigid motion invariant feature - the trace of a 3 x 3 "feature matrix" related to the moment of inertia tensor. This trace can be computed at every point of a moving surface and provides a quantitative measure of the local shape of the surface. Based on the feature's conservation along the trajectory of every moving point, a 3-D Flow Constraint Equation is formulated and solved for the velocity field. The robustness of the feature in presence of noise and discontinuity is analyzed. Index Terms - Motion analysis, optical flow, rigid motion invariants, stereo, 3-D flow, 3-D motion detection.

Conventional computer vision methods for determining a robot's end-effector motion based on sensory data needs sensor calibration (e.g., camera calibration) and sensor-to-hand calibration (e.g., hand-eye calibration). This involves many computations and even some difficulties, especially when different kinds of sensors are involved. In this correspondence, we present a neural network approach to the motion determination problem without any calibration. Two kinds of sensory data, namely, camera images and laser range data, are used as the input to a multilayer feedforward network to associate the direct transformation from the sensory data to the required motions. This provides a practical sensor fusion method. Using a recursive motion strategy and in terms of a network correction, we relax the requirement for the exactness of the learned transformation. Another important feature of our work is that the goal position can be changed without having to do network retraining. Experimental results show the effectiveness of our method. Index Terms - Cameras, laser range finders, neural networks, visual servoing.

To discriminate among all possible diagnoses using Hou's theory of measurement in diagnosis from first principles, one has to derive all minimal conflict sets from a known conflict set. However, the result derived from Hou's method depends on the order of node generation in CS-trees. We develop a derivation method with mark set to overcome this drawback of Hou's method. We also show that our method is more efficient in the sense that no redundant tests have to be done. An enhancement to our method with the aid of extra information is presented. Finally, a discussion on top-down and bottom-up derivations is given. Index Terms - Measurement, minimal hitting set, observation, system description.

Based on human retinal sampling distributions and eye movements, a sequential resolution image preprocessor is developed. Combined with a nearest neighbor classifier, this preprocessor provides an efficient image classification method, the sequential resolution nearest neighbor (SRNN) classifier. The human eye has a typical fixation sequence that exploits the nonuniform sampling distribution of its retina. If the retinal resolution is not sufficient to identify an object, the eye moves in such a way that the projection of the object falls onto a retinal region with a higher sampling density. Similarly, the SRNN classifier uses a sequence of increasing resolutions until a final class decision is made. Experimental results on texture segmentation show that the preprocessor used in the SRNN classifier is considerably faster than traditional multiresolution algorithms which use all the available resolution levels to analyze the input data. Index Terms - Fixation point, fovea, image classification, multiresolution analysis.

Standard least-squares (LS) methods for pose estimation of objects are sensitive to outliers which can occur due to mismatches. Even a single mismatch can severely distort the estimated pose. This paper describes a least-median of squares (LMedS) approach to estimating pose using point matches. It is both robust (resistant to up to 50% outliers) and efficient (linear in the number of points). The basic algorithm is then extended to improve performance in the presence of two types of noise: 1) type I which perturbs all data values by small amounts (e.g., Gaussian) and 2) type II which can corrupt a few data values by large amounts. Index Terms - Least median squares, robust pose estimation.

Two statistical language models have been investigated on their effectiveness in upgrading the accuracy of a Chinese character recognizer. The baseline model is one of lexical analytic nature which segments a sequence of character images according to the maximum matching of words with consideration of word binding forces. A model of bigram statistics of word-classes is then investigated and compared against the baseline model in terms of recognition rate improvement on the image recognizer. On the average, the baseline language model improves the recognition rate by about 7% while the bigram statistics model upgrades it by about 10%. Index Terms - Chinese character recognizer, handwritten Chinese character recognizer, language model, statistic language model.

This paper shows that in edge detection the regularization factor [Alpha] is a better scale parameter than the standard deviation ([Sigma]) of the Gaussian pre-filter. The [Alpha] scale space, which exhibits the evolutionary behavior of an edge in various scales, is the basis for the design of a multiscale edge detector (MRCBS). In MRCBS, the scale is determined adaptively according to the local noise level; the thresholds which control the amount of edge details are adjusted according to the scale; and the anisotropic diffusion is applied in the finest scale to further suppress noise. Index Terms - Adaptive signal processing, curve fitting, diffusion equation, image edge analysis, spline function.

An artificial neural network (ANN) is employed for sorting a sequence of real elements in monotonic (descending or ascending) order. Although inspired by harmony theory (HT), whereby the same construction as for the HT ANN is followed, the proposed ANN differs in the mode of operation, namely the obliteration of the consensus (harmony) function, the circumvention of simulated annealing as a means of settling to a solution, the simplification of the activation updating of the nodes of the upper layer, the clamping of the nodes of the lower layer, the gradual shrinking of the ANN and the use of an automatic termination criterion. The creation of the sorted sequence is progressive, whereby at most as many network updates are required as there are elements in the sequence. Ties between elements are resolved by simultaneous activation of the corresponding nodes. Finally, the min and max problems are solved in a single network update. Index Terms - Artificial neural networks, harmony theory, sorting.

In this paper, we have presented an alternate ANN structure called functional link ANN (FLANN) for nonlinear dynamic system identification using the popular back propagation algorithm. In contrast to a feed forward ANN structure, i.e., a multilayer perceptron (MLP), the FLANN is basically a single layer structure in which nonlinearity is introduced by enhancing the input pattern with nonlinear functional expansion. With proper choice of functional expansion in a FLANN, this network performs as good as and in some cases even better than the MLP structure for the problem of nonlinear system identification. Index Terms - Artificial neural networks, computational complexity, nonlinear dynamic system identification.

The motion detection problem occurs frequently in many applications connected with computer vision. Researchers have studied motion detection based on naturally occurring biological circuits for over a century. In this paper, we propose and analyze a motion detection circuit which is based on nerve membrane conduction. It consists of two unidirectional neural networks connected in an opposing fashion. Volterra input - output (I-O) models are then derived for the network so that velocity estimation can be cast as a parameter estimation problem. The technique is demonstrated through simulation. Index Terms - Image motion analysis, neural network architecture, nonlinear systems, parameter estimation, Volterra series.

We discuss a novel model for analyzing the working of Genetic Algorithms (GA's), when the objective function is a function of unitation. The model is exact (not approximate), and is valid for infinite populations. Functions of unitation depend only on the number of 1's in any string. Hence, we only need to model the variations in the distribution of strings with respect to the number of 1's in the strings. We introduce the notion of a Binomial Distributed Population (BDP) as the building block of our model, and we show that the effect of uniform crossover on BDP's is to generate two other BDP's. We demonstrate that a population with any general distribution may be decomposed into several BDP's. We also show that a general multipoint crossover may be considered as a composition of several uniform crossovers. Based on these results, the effects of mutation and crossover on the distribution of strings have been characterized, and the model has been defined. GASIM - a Genetic Algorithm Simulator for functions of unitation - has been implemented based on the model, and the exactness of the results obtained from GASIM has been verified using actual Genetic Algorithm runs. The time complexity of the GA simulator derived from the model is O([l.sup.3]) (where l is the string length), a significant improvement over previous models with exponential time complexities. As an application of GASIM, we have analyzed the effect of crossover rate on deception in trap functions, a class of deceptive functions of unitation. We have obtained interesting results - we are led to believe that increasing values of [p.sub.c], the crossover rate, increase the probability of the GA converging to the local optimum of the trap function.

Marshall McLuhan's theory of sensory functions shows that human sensory experiences influence the way people react to and interact with events. It aims to demonstrate the role of technological innovations in human sensory functions and the cultural use of environment to activate the senses and intensify some particular senses. McLuhan's theory, however, cannot effectively explain and predict the extent to which these factors are affective, since the model makes certain assumptions that cannot be substantiated by the procedures taken.

In this paper, a model which combines relational databases with self-processing networks is proposed in order to improve the performance of very large databases. The proposed model uses an approach which is radically different from all other distributed database models, where each computer processes a portion of the database. In the self-processing network model, the network structure which consists of nodes and connections, captures the data and the relationships by assigning them unique, connected control, and data nodes. The network activity is the mechanism that performs the relational algebra operations. No data transmission is needed, and since data nodes are common to all the relations, integrity and elimination of data redundancy are achieved. An extension of the model, by interconnecting the data nodes via weighted links, provides us with properties that are embedded in neural networks, such as fuzziness and learning. Index Terms - Relational databases, self-processing network.

The expressions of "excluded middle" and "crisp contradiction" are reexamined starting with their original linguistic expressions which are first restated in propositional and then predicate forms. It is shown that, in order to generalize the truth tables and hence the normal forms, the membership assignments in predicate expressions must be separated from their truth qualification. In two-valued logic, there is no need to separate them from each other due to reductionist Aristotalean dichotomy. Whereas, in infinite (fuzzy) valued set and logic, the separation of membership assignments from their truth qualification forms the bases of a new reconstruction of the truth tables. The results obtained from these extended truth tables are reducible to their Boolean equivalents under the axioms of Boolean theory. Whereas, in fuzzy set and logic theory, we obtain a richer and more complex interpretations of the "fuzzy middle" and "fuzzy contradiction." Index Terms - Crisp contradiction, excluded middle, extended truth tables, fuzzy contradiction, fuzzy middle, fuzzy normal forms, linguistic expressions, propositional and predicate forms, reaffirmation, renegation.

Due to their powerful optimization property, genetic algorithms (GA's) are currently being investigated for the development of adaptive or self-tuning fuzzy logic control systems. This paper presents a neuro-fuzzy logic controller (NFLC) where all of its parameters can be tuned simultaneously by GA. The structure of the controller is based on the radial basis function neural network (RBF) with Gaussian membership functions. The NFLC tuned by GA can somewhat eliminate laborious design steps such as manual tuning of the membership functions and selection of the fuzzy rules. The GA implementation incorporates dynamic crossover and mutation probabilistic rates for faster convergence. A flexible position coding strategy of the NFLC parameters is also implemented to obtain near optimal solutions. The performance of the proposed controller is compared with a conventional fuzzy controller and a PID controller tuned by GA. Simulation results show that the proposed controller offers encouraging advantages and has better performance. Index Terms - Auto-tuning, dynamic crossover, genetic algorithms, neuro-fuzzy, radial basis functions.

Application of artificial neural networks (ANN's) to adaptive channel equalization in a digital communication system with 4-QAM signal constellation is reported in this paper. A novel computationally efficient single layer functional link ANN (FLANN) is proposed for this purpose. This network has a simple structure in which the nonlinearity is introduced by functional expansion of the input pattern by trigonometric polynomials. Because of input pattern enhancement, the FLANN is capable of forming arbitrarily nonlinear decision boundaries and can perform complex pattern classification tasks. Considering channel equalization as a nonlinear classification problem, the FLANN has been utilized for nonlinear channel equalization. The performance of the FLANN is compared with two other ANN structures [a multilayer perceptron (MLP) and a polynomial perceptron network (PPN)] along with a conventional linear LMS-based equalizer for different linear and nonlinear channel models. The effect of eigenvalue ratio (EVR) of input correlation matrix on the equalizer performance has been studied. The comparison of computational complexity involved for the three ANN structures is also provided. Index Terms - Functional link artificial neural networks, multilayer perceptron, nonlinear channel equalization, polynomial perceptron, QAM signals.

The purpose of this paper is to design and rigorously analyze a tracking controller, based on a dynamic neural network model for unknown but affine in the control, multi input nonlinear dynamical systems. Lyapunov stability theory is used to guarantee a uniform ultimate boundedness property for the tracking error, as well as of all other signals in the closed loop. The controller derived is smooth. No a priori knowledge of an upper bound on the "optimal" weights and modeling errors is required. Simulation studies are used, to illustrate and clarify the theoretical results. Index Terms - Adaptive control, neural networks, nonlinear systems.