Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Bücher
Zeitschriften
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
SLX-Digitalkonferenz: Zukunftswerkstatt 2024

Mehr Umsatz mit Top-Kontakten

Am 7. Mai erfahren Sie, wie Vertriebsteams Leadgenerierung, Leadnurturing und Leadstrategien effizient steuern können.
Erweiterte Suche
Anmelden
nach oben

2007 | Buch

Hybridization Schemes in Architectures of Computational Intelligence Kapitel lesen Erstes Kapitel lesen

Hybrid Intelligent Systems

Analysis and Design

herausgegeben von: Oscar Castillo, Patricia Melin, Janusz Kacprzyk, Witold Pedrycz

Verlag: Springer Berlin Heidelberg

Buchreihe : Studies in Fuzziness and Soft Computing

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

We describe in this book, new methods for analysis and design of hybrid intelligent systems using soft computing techniques. Soft Computing (SC) consists of several computing paradigms, including fuzzy logic, neural networks, and genetic algorithms, which can be used to produce powerful hybrid intelligent systems for solving problems in pattern recognition, time series prediction, intelligent control, robotics and automation. Hybrid int- ligent systems that combine several SC techniques are needed due to the complexity and high dimensionality of real-world problems. Hybrid int- ligent systems can have different architectures, which have an impact on the efficiency and accuracy of these systems, for this reason it is very - portant to optimize architecture design. The architectures can combine, in different ways, neural networks, fuzzy logic and genetic algorithms, to achieve the ultimate goal of pattern recognition, time series prediction, - telligent control, or other application areas. This book is intended to be a major reference for scientists and en- neers interested in applying new computational and mathematical tools to design hybrid intelligent systems. This book can also be used as a textbook or major reference for graduate courses like the following: soft computing, intelligent pattern recognition, computer vision, applied artificial intel- gence, and similar ones. We consider that this book can also be used to get novel ideas for new lines of research, or to continue the lines of research proposed by the authors of the book.

Inhaltsverzeichnis

Frontmatter

Theory

Frontmatter
Hybridization Schemes in Architectures of Computational Intelligence
Abstract
While the essence of Computational Intelligence hinges profoundly on the symbiotic use of their underlying technologies (viz. neurocomputing, granular computing, and predominantly fuzzy sets, and evolutionary optimization), there are several other equally promising development avenues where a hybrid usage of the underlying technologies is worth pursuing. In this study, we concentrate on the hybrid concepts and constructs available within the realm of Granular Computing (GC). Given the highly diversified landscape of GC, we discuss main directions of forming hybrid structures involving individual technologies of information granulation, elaborate on the fundamental communication, interoperability, and orthogonality issues and propose some general ways of building hybrid constructs of GC which are of immediate interest to system modeling realized in the realm of Computational Intelligence. We also shed light on the central role of the concepts of information granularity, information granules and ensuing hybrid constructs. Furthermore we emphasize a role of hierarchical modeling that is directly supported by stratified aspect of information granules formed at nested levels of specificity. The central issue of human-centricity of such models is also highlighted.
Witold Pedrycz
ChapBoltzmann Machines Learning Using High Order Decimation
Abstract
Boltzmann Machines are recurrent and stochastic neural networks that can learn and reproduce probability distributions. This feature has a serious drawback in the exhaustive computational cost involved. In this context, decimation was introduced as a way to overcome this problem, as it builds a smaller network that is able to reproduce exactly the quantities required to update the weights during learning. Decimation techniques developed can only be used in sparsely connected Boltzmann Machines with stringent constraints on the connections between the units. In this work, decimation is extended to any Boltzmann Machine with no restrictions on connections or topology. This is achieved introducing high order weights, which incorporate additional degrees of freedom.
Enric Farguell, Ferran Mazzanti, Eduardo Gomez-Ramirez
Evolutionary Optimization of a Wiener Model
Abstract
There exists no standard method for obtaining a nonlinear inputoutput model using external dynamic approach. In this work, we are using an evolutionary optimization method for estimating the parameters of an NFIR model using the Wiener model structure. Specifically we are using a Breeder Genetic Algorithm (BGA) with fuzzy recombination for performing the optimization work. We selected the BGA since it uses real parameters (it does not require any string coding), which can be manipulated directly by the recombination and mutation operators. For training the system we used amplitude modulated pseudo random binary signal (APRBS). The adaptive system was tested using sinusoidal signals.
Oscar Montiel, Oscar Castillo, Patricia Melin, Roberto Sepúlveda
Synchronization of Chaotic Neural Networks: A Generalized Hamiltonian Systems Approach
Abstract
to synchronize chaotic neural networks unidirectionally coupled. Synchronization is thus between the master and the slave networks with the slave network being given by an observer. In this paper, we use a Generalized Hamiltonian forms approach
C. Posadas-Castillo, C. Cruz-Hernández, D. López-Mancilla
Mediative Fuzzy Logic: A Novel Approach for Handling Contradictory Knowledge
Abstract
In this paper we are proposing a novel fuzzy method that can handle imperfect knowledge in a broader way than Intuitionistic fuzzy logic does (IFL). This fuzzy method can manage non-contradictory, doubtful, and contradictory information provided by experts, providing a mediated solution, so we called it Mediative Fuzzy Logic (MFL). We are comparing results of MFL, with IFL and traditional Fuzzy logic (FL).
Oscar Montiel, Oscar Castillo

Intelligent Control Applications

Frontmatter
Direct and Indirect Adaptive Neural Control of Nonlinear Systems
Abstract
A comparative study of various control systems using neural networks is done. The paper proposes to use a Recurrent Trainable Neural Network (RTNN) identifier with backpropagation method of learning. Two methods of adaptive neural control with integral plus state action are applied - an indirect and a direct trajectory tracking control. The first one is the indirect Sliding Mode Control (SMC) with I-term where the SMC is resolved using states and parameters identified by RTNN. The second one is the direct adaptive control with I-term where the adaptive control is resolved by a RTNN controller. The good tracking abilities of both methods are confirmed by simulation results obtained using a MIMO mechanical plant and a 1-DOF mechanical system with friction plant model. The results show that both control schemes could compensate constant offsets and that - without I- term did not.
Ieroham Baruch
Simple Tuning of Fuzzy Controllers
Abstract
The number of applications in the industry using the PID controllers is bigger than fuzzy controllers. One reason is the problem of the tuning, because it implies the handling of a great quantity of variables like: the shape, number and ranges of the membership functions, the percentage of overlap among them and the design of the rule base. The problem is more complicated when it is necessary to control multivariable systems due that the number of parameters. The importance of the tuning problem implies to obtain fuzzy system that decrease the settling time of the processes in which it is applied, or in some cases, the settling time must be fixed to some specific value. In this work a very simple algorithm is presented for the tuning of a fuzzy controller using only one variable to adjust the performance of the system. The results are based on the relation that exists between the shape of the membership functions and the settling time. Some simulations are presented to exemplified the algorithm proposed.
Eduardo Gómez-Ramírez
From Type-1 to Type-2 Fuzzy Logic Control: A Stability and Robustness Study
Abstract
Stability is one of the more important aspects in the traditional knowledge of Automatic Control. Type-2 Fuzzy Logic is an emerging and promising area for achieving Intelligent Control (in this case, Fuzzy Control). In this work we use the Fuzzy Lyapunov Synthesis as proposed by Margaliot [11] to build a Lyapunov Stable Type-1 Fuzzy Logic Control System, and then we make an extension from a Type-1 to a Type-2 Fuzzy Logic Control System, ensuring the stability on the control system and proving the robustness of the correponding fuzzy controller.
Nohé Cázarez, Oscar Castillo, Luís Aguilar, Selene Cárdenas
A Comparative Study of Controllers Using Type-2 and Type-1 Fuzzy Logic
Abstract
applications. The use of new methods for handling incomplete information is of fundamental importance in engineering applications. This paper deals with the design of controllers using type-2 fuzzy logic for minimizing the effects of uncertainty produced by the instrumentation elements. We simulated type-1 and type-2 fuzzy logic controllers to perform a comparative analysis of the systems’ response, in the presence of uncertainty. Uncertainty is an inherent part in controllers used for real-world
Roberto Sepulveda, Patricia Melin
Evolutionary Computing for Topology Optimization of Type-2 Fuzzy Controllers
Abstract
We describe in this paper the use of hierarchical genetic algorithms for fuzzy system optimization in intelligent control. In particular, we consider the problem of optimizing the number of rules and membership functions using an evolutionary approach. The hierarchical genetic algorithm enables the optimization of the fuzzy system design for a particular application. We illustrate the approach with the case of intelligent control in a medical application. Simulation results for this application show that we are able to find an optimal set of rules and membership functions for the fuzzy system.
Oscar Castillo, Gabriel Huesca, Fevrier Valdez

Robotic Applications

Frontmatter
Decision Trees and CBR for the Navigation System of a CNN-based Autonomous Robot
Abstract
In this paper we present a navigation system based on decision trees and CBR (Case-Based reasoning) to guide an autonomous robot. The robot has only real-time visual feedback, and the image processing is performed by CNNs to take advantage of the parallel computation. We successfully tested the system on a SW simulator.
Giovanni Egidio Pazienza, Elisabet Golobardes-Ribé, Xavier Vilasís-Cardona, Marco Balsi
Intelligent Agents in Distributed Fault Tolerant Systems
Abstract
Intelligent Agents have originated a lot of discussion about what they are, and how they are different from general programs. We describe in this paper a new paradigm for intelligent agents. This paradigm helped us deal with failures in an independent and efficient way. We proposed three types of agents to treat the system in a hierarchic way. A new way to visualize fault tolerant systems (FTS) is proposed, in this paper with the incorporation of intelligent agents, which as they grow and specialized create the Multi-Agent System (MAS). The MAS contains a diversified range of agents, which depending on the perspective will be specialized or evolutionary (from our initially proposal) they will be specialized for the detection and possible solution of errors that appear in an FTS). The initial structure of the agent is proposed in [1] and it is called a reflected agent with an internal state and in the Method MeCSMA [2].
Arnulfo Alanis Garza, Juan José Serrano, Rafael Ors Carot, José Mario, García Valdez
Genetic Path Planning with Fuzzy Logic Adaptation for Rovers Traversing Rough Terrain
Abstract
The paper develops a genetic algorithm approach to path planning for a mobile robot operating in rough environments. Path planning consists of a description of the environment using a fuzzy logic framework, and a two-stage planner. A global planner determines the path that optimizes a combination of terrain roughness and path curvature. A local planner uses sensory information, and in case of detection of previously unknown and unaccounted for obstacles, performs an on-line planning to get around the newly discovered obstacle. The adaptation of the genetic operators is achieved by adjusting the probabilities of the genetic operators based on a diversity measure of the population and traversability measure of the path. Path planning for an articulate rover in a rugged Mars terrain is presented to demonstrate the effectiveness of the proposed path planner.
Mahmoud Tarokh
Chattering Attenuation Using Linear-in-the-Parameter Neural Nets in Variable Structure Control of Robot Manipulators with Friction
Abstract
Variable structure control is a recognized method to stabilize mechanical systems with friction. Friction produces non linear phenomena, such as tracking errors, limit cycles, and undesired stick-slip motion, degrading the performance of the closed-loop system. The main drawback of variable structure control is the presence of chattering, which is not suitable in mechanical systems. In this paper, we design a variable structure controller complemented with Linear-in-the-Parameter neural nets to attenuate chattering. Experimental validation applied to a three degree of freedom robot mechanical manipulator is shown to support the results.
Ricardo Guerra, Luis T. Aguilar, Leonardo Acho
Tracking Control for a Unicycle Mobile Robot Using a Fuzzy Logic Controller
Abstract
We develop a tracking controller for the dynamic model of unicycle mobile robot by integrating a kinematic controller and a torque controller based on Fuzzy Logic Theory. Computer simulations are presented confirming the performance of the tracking controller and its application to different navigation problems.
Selene L. Cárdenas, Oscar Castillo, Luis T. Aguilar, Nohé Cázarez
Intelligent Control and Planning of Autonomous Algorithms Mobile Robots Using Fuzzy Logic and Genetic
Abstract
problem of Offline Point-to-Point Autonomous Mobile Robot Path Planning. The problem consist of generating “valid” paths or trajectories, for an Holonomic Robot to use to move from a starting position to a destination across a flat map of a terrain, represented by a two dimensional grid, with obstacles and dangerous ground that the Robot must evade. This means that the GA optimizes possible paths based on two criteria: length and difficulty. This paper describes the use of a Genetic Algorithm (GA) for the
Julian Garibaldi, Azucena Barreras, Oscar Castillo

Pattern Recognition Applications

Frontmatter
The Role of Neural Networks in the Interpretation of Antique Handwritten Documents
Abstract
The need for accessing information through the web and other kind of distributed media makes it mandatory to convert almost every kind of document to a digital representation. However, there are many documents that were created long time ago and currently, in the best cases, only scanned images of them are available, when a digital transcription of their content is needed. For such reason, libraries across the world are looking for automatic OCR systems able to transcript that kind of documents. In this chapter we describe how Artificial Neural Networks can be useful in the design of an Optical Character Recognizer able to transcript handwritten and printed old documents. The properties of Neural Networks allow this OCR to have the ability to adapt to the styles of handwritten or antique fonts. Advances with two prototype parts of such OCR are presented.
Pilar Gómez-Gil, Guillermo De los Santos-Torres, Jorge Navarrete-García, Manuel Ramírez-Cortés
Reasoning Object Recognition Using Fuzzy Inferential
Abstract
This paper introduces a vision-based pattern recognition scheme for the identification of very high tolerances of manufactured industrial objects. An image-forming device is developed for the generation and the capture of images/silhouettes of the components. A simple but effective feature extraction algorithm is employed to produce distinguishable features of the components in question. Radial basis function (RBF) based membership functions are used as classifiers for the pattern classification. For the decision making process, a fuzzy logic based inferential reasoning algorithm is implemented for the approximate reasoning scheme.
Thompson Sarkodie-Gyan
The Fuzzy Sugeno Integral as a Decision Operator in the Recognition of Images with Modular Neural Networks
Abstract
In a previous paper we presented the implementation of the Fuzzy Sugeno Integral formulas developed with Matlab 6.5™. The programs are now included in a System called “Herramientas Multired” (“hmr”). In this paper we will review an example of modular neural network for image recognition, using images divided in four parts. The Fuzzy Sugeno Integral was used to make a final decision for pattern recognition.
Olivia Mendoza Duarte, Patricia Melin
Modular Neural Networks and Fuzzy Sugeno Integral for Pattern Recognition: The Case of Human Face and Fingerprint
Abstract
We describe in this paper a new approach for pattern recognition using modular neural networks with a fuzzy logic method for response integration. We proposed a new architecture for modular neural networks for achieving pattern recognition in the particular case of human faces and fingerprints. Also, the method for achieving response integration is based on the fuzzy Sugeno integral with some modifications. Response integration is required to combine the outputs of all the modules in the modular network. We have applied the new approach for fingerprint and face recognition with a real database from students of our institution.
Patricia Melin, Claudia Gonzalez, Diana Bravo, Felma Gonzalez, Gabriela Martinez

Time Series and Diagnosis

Frontmatter
Optimal Training for Associative Memories: Application to Fault Diagnosis in Fossil Electric Power Plants
Abstract
In this chapter, the authors discuss a new synthesis approach to train associative memories, based on recurrent neural networks. They propose to update the weight vector as the optimal solution of a linear combination of support patterns. The proposed training algorithm maximizes the margin between the training patterns and the decision boundary. This algorithm is applied to the synthesis of an associative memory, for fault diagnosis in fossil electric power plants. The scheme is evaluated via a full scale simulator to diagnose the main faults occurred in this kind of power plants.
Jose A. Ruz-Hernandez, Edgar N. Sanchez, Dionisio A. Suarez
Acceleration Output Prediction of Buildings Using a Polynomial Artificial Neural Network
Abstract
Severe earthquake motions could make civil structures to undergo hysteretic cycles and crack or yield their resistant elements. The present research proposes the use of a polynomial artificial neural network to identify and predict, on-line, the behavior of such nonlinear systems. Predictions are carried out first on theoretical hysteretic models and later using two real seismic records acquired on a 24-story concrete building in Mexico City. Only two cycles of movement are needed for the identification process and the results show fair prediction of the acceleration output.
Francisco J. Rivero-Angeles, Eduardo Gomez-Ramirez
Time Series Forecasting of Tomato Prices and Processing in Parallel in Mexico Using Modular Neural Networks
Abstract
In this paper we describe the concepts of Time Series, Neural Networks, Modular Neural Networks, and Parallelism. Modular Neural Networks and Parallel Processing for Time Series Forecasting of Tomato Prices in Mexico are described in this paper. A particular modular neural network architecture implemented in parallel was used. Simulation results with the modular neural network approach for this application are very good.
Ileana Leal, Patricia Melin
Modular Neural Networks with Fuzzy Sugeno Integration Applied to Time Series Prediction
Abstract
We describe in this paper the application of several neural network architectures to the problem of simulating and predicting the dynamic behavior of complex economic time series. We use several neural network models and training algorithms to compare the results and decide at the end, which one is best for this application. We also compare the simulation results with the traditional approach of using a statistical model. In this case, we use real time series of prices of consumer goods to test our models. Real prices of tomato and green onion in the U.S. show complex fluctuations in time and are very complicated to predict with traditional statistical approaches.
Patricia Melin, Valente Ochoa, Luis Valenzuela, Gabriela Torres, Daniel Clemente
On Linguistic Summaries of Time Series Using a Fuzzy Quantifier Based Aggregation via the Sugeno Integral
Abstract
We propose and advocate the use of linguistic summaries as descriptions of trends in time series data. We consider two general types of such summaries: summaries based on frequence and summaries based on duration. We employ the concept of a linguistic database summary due to Yager. To account for a specificity of time series data summarization we employ the Sugeno integrals for linguistic quantifier based aggregation.
Janusz Kacprzyk, Sławomir Zadrożny, Anna Wilbik
Metadaten
Titel
Hybrid Intelligent Systems
herausgegeben von
Oscar Castillo
Patricia Melin
Janusz Kacprzyk
Witold Pedrycz
Copyright-Jahr
2007
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-540-37421-3
Print ISBN
978-3-540-37419-0
DOI
https://doi.org/10.1007/978-3-540-37421-3

Premium Partner

    Bildnachweise