Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top

2009 | Book

Optimization of Membership Functions of a Fuzzy Logic Controller for an Autonomous Wheeled Mobile Robot Using Ant Colony Optimization Read chapter Read first chapter

Evolutionary Design of Intelligent Systems in Modeling, Simulation and Control

Editors: Oscar Castillo, Witold Pedrycz, Janusz Kacprzyk

Publisher: Springer Berlin Heidelberg

Book Series : Studies in Computational Intelligence

Part of: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Login to get access
insite
SEARCH

About this book

We describe in this book, new methods for evolutionary design of intelligent s- tems using soft computing and their applications in modeling, simulation and c- trol. Soft Computing (SC) consists of several intelligent computing paradigms, including fuzzy logic, neural networks, and evolutionary algorithms, which can be used to produce powerful hybrid intelligent systems. The book is organized in four main parts, which contain a group of papers around a similar subject. The first part consists of papers with the main theme of evolutionary design of fuzzy systems in intelligent control, which consists of papers that propose new methods for designing and optimizing intelligent controllers for different applications. The second part c- tains papers with the main theme of evolutionary design of intelligent systems for pattern recognition applications, which are basically papers using evolutionary al- rithms for optimizing modular neural networks with fuzzy systems for response - tegration, for achieving pattern recognition in different applications. The third part contains papers with the themes of models for learning and social simulation, which are papers that apply intelligent systems to the problems of designing learning - jects and social agents. The fourth part contains papers that deal with intelligent s- tems in robotics applications and hardware implementations. In the part of Intelligent Control there are 5 papers that describe different c- tributions on evolutionary optimization of fuzzy systems in intelligent control. The first paper, by Ricardo Martinez-Marroquin et al.

Table of Contents

Frontmatter

Intelligent Control

Frontmatter
Optimization of Membership Functions of a Fuzzy Logic Controller for an Autonomous Wheeled Mobile Robot Using Ant Colony Optimization
Abstract
In this paper we describe the application of a Simple ACO (S-ACO) as a method of optimization for membership functions’ parameters of a fuzzy logic controller (FLC) in order to find the optimal intelligent controller for an Autonomous Wheeled Mobile Robot. Simulation results show that ACO outperforms a GA in the optimization of FLCs for an autonomous mobile robot.
Ricardo Martínez-Marroquín, Oscar Castillo, José Soria
Evolutionary Optimization of Type-2 Fuzzy Logic Systems Applied to Linear Plants
Abstract
We describe a different kind of evolutionary methods to optimize a type-2 fuzzy logic controller (FLC) applied to linear plants. The evolutionary method used is a genetic algorithm to find the optimal FLC for the plant control. The plant receives a linear signal of input controlled by an optimized FLC, obtaining as result the control and the stability of the plant. Simulations results were made in Simulink showing the effectiveness of the proposal.
Ricardo Martinez, Oscar Castillo, Luis T. Aguilar, Antonio Rodriguez
Multi-Agent System with Fuzzy Logic Control for Autonomous Mobile Robots in Known Environments
Abstract
This paper describes the development of a Multi-Agent System (MAS), which is supported with fuzzy logic (to control the robots movements in a reactive path) and vision, which controls an autonomous mobile robot to exit a maze. The research consists of two stages. In the first stage the problem is to be able to make the robot exit a maze, the mobile robot is positioned at the entrance (point A) and should reach an output (B). It should be noted that we are working with a NXT Lego MINDSTORMS robot. In its second phase the problem is to make the robot search for a recognized object, for this, a camera is used to capture images, which will be processed with vision techniques, for their identification, and after that, the SMA takes the decision to evade or take the object as appropriate.
Cinthya Solano-Aragón, Arnulfo Alanis
Hybrid Interval Type-1 Non-singleton Type-2 Fuzzy Logic Systems Are Type-2 Adaptive Neuro-fuzzy Inference Systems
Abstract
This article presents a new learning methodology based on a hybrid algorithm for interval type-1 non-singleton type-2 TSK fuzzy logic systems (FLS). Using input-output data pairs during the forward pass of the training process, the interval type-1 non-singleton type-2 TSK FLS output is calculated and the consequent parameters are estimated by the recursive least-squares (RLS) method. In the backward pass, the error propagates backward, and the antecedent parameters are estimated by the back-propagation (BP) method. The proposed hybrid methodology was used to construct an interval type-1 non-singleton type-2 TSK fuzzy model capable of approximating the behavior of the steel strip temperature as it is being rolled in an industrial Hot Strip Mill (HSM) and used to predict the transfer bar surface temperature at finishing Scale Breaker (SB) entry zone. Comparative results show the performance of the hybrid learning method (RLS-BP) against the only BP learning.
Gerardo M. Mendez, Ma. De Los Angeles Hernandez
Centralized Direct and Indirect Neural Control of Distributed Parameter Systems
Abstract
The paper proposed to use a Recurrent Neural Network Model (RNNM) for centralized modeling, identification and direct adaptive control of an anaerobic digestion bioprocess, carried out in a fixed bed and a recirculation tank of a wastewater treatment system. The analytical model of the digestion bioprocess represented a distributed parameter system, which is reduced to a lumped system using the orthogonal collocation method, applied in three collocation points plus the recirculation tank. The RNNM learning algorithm is the dynamic backpropagation one. The graphical simulation results of the distributed plant direct and indirect adaptive neural control system, exhibited good convergence and precise reference tracking, outperforming the optimal control.
Ieroham S. Baruch, Rosalba Galvan-Guerra

Pattern Recognition

Frontmatter
An Ensemble Neural Network Architecture with Fuzzy Response Integration for Complex Time Series Prediction
Abstract
In this paper we describe the application of an architecture for an ensemble neural network for Complex Time Series Prediction. The times series we are considering are: the Mackey-Glass, Dow Jones and Mexican Stock Exchange and we show the results of a set of trainings with the ensemble neural network, and its integration with the methods of average, weighted average and Fuzzy Integration. Simulation results show very good prediction of the ensemble neural network with fuzzy logic integration.
Martha Pulido, Alejandra Mancilla, Patricia Melin
Optimization of Fuzzy Response Integrators in Modular Neural Networks with Hierarchical Genetic Algorithms: The Case of Face, Fingerprint and Voice Recognition
Abstract
In this paper we describe the development of Fuzzy Response Integrators of Modular Neural Networks (MNN) for Face, Fingerprint and Voice Recognition, and their Optimization with a Hierarchical Genetic Algorithm (HGA). The optimization of the integrators consists of optimizing their membership functions, fuzzy rules, type of model (Mamdani or Sugeno), type of fuzzy logic (type-1 or type-2). The MNN architecture consists of three modules; face, fingerprint and voice. Each of the modules is divided again into three sub modules. The same information is used as input to train the sub modules. Once we have trained and tested the MNN modules, we proceed to integrate these modules with an optimized fuzzy integrator. In this paper we show that using a HGA as an optimization technique for the fuzzy integrators is a good option to solve MNN integration problems.
Ricardo Muñoz, Oscar Castillo, Patricia Melin
Modular Neural Network with Fuzzy Integration of Responses for Face Recognition
Abstract
This paper presents a modular neural network with fuzzy integration of responses for face recognition. We describe its architecture and simulation results using the ORL database. We show results from different integrators, such as the Gating Network, fuzzy Sugeno integrals and type-1 fuzzy systems. We also show that the results with type-1 fuzzy systems are good, but we decided to optimize this fuzzy system with genetic algorithms (MFs parameters and fuzzy rules) to improve the results.
Erika Ayala, Miguel Lopez, Patricia Melin
A Modular Neural Network with Fuzzy Response Integration for Person Identification Using Biometric Measures
Abstract
This paper describes an intelligent system for person identification with biometric measures such as signature, fingerprint and face. We describe the neural network architectures used to achieve person identification based on the biometrics measures. Simulation results show that the proposed method provides good recognition. Fuzzy integration of the three modules is tested on a single computer and also in a distributed environment.
Magdalena Serrano, Patricia Melin
Signature Recognition with a Hybrid Approach Combining Modular Neural Networks and Fuzzy Logic for Response Integration
Abstract
This chapter describes a modular neural network (MNN) with fuzzy integration for the problem of signature recognition. Currently, biometric identification has gained a great deal of research interest within the pattern recognition community. For instance, many attempts have been made in order to automate the process of identifying a person’s handwritten signature; however this problem has proven to be a very difficult task. In this work, we propose a MNN that has three separate modules, each using different image features as input, these are: edges, wavelet coefficients, and the Hough transform matrix. Then, the outputs from each of these modules are combined using a Sugeno fuzzy integral and a fuzzy inference system. The experimental results obtained using a database of 30 individual’s shows that the modular architecture can achieve a very high 99.33% recognition accuracy with a test set of 150 images. Therefore, we conclude that the proposed architecture provides a suitable platform to build a signature recognition system. Furthermore we consider the verification of signatures as false acceptance, false rejection and error recognition of the MNN.
Mónica Beltrán, Patricia Melin, Leonardo Trujillo

Learning and Social Simulation

Frontmatter
A Hybrid Recommender System Architecture for Learning Objects
Abstract
In this paper we present the architecture of a hybrid recommender system to support an adaptive hypermedia educational (AHE) system. Currently the instructor (using fuzzy rules) specifies the sequence in which learning objects are presented to students. The instructor can also give students a chance to choose from a pool of objects and helps them make their selection by assigning to each object a recommendation rating based on the student’s profile. We propose a hybrid recommender system that uses collaborative filtering techniques together with fuzzy inference systems to provide recommendations, considering the instructor’s experience as well as the ratings given by similar students.
Mario García-Valdez, Brunett Parra
TA-Fuzzy Semantic Networks for Interaction Representation in Social Simulation
Abstract
The need to model interactions between people of different cultures, religions and ethnic groups is evident. In Social Simulation, the combination of Artificial Intelligence andMulti-agent Systems has proven to be a good tool for modeling social groups, however much remains to achieve a model which represents a society with differences between individuals. Our proposal is to combine fuzzy logic, semantic networks and transactional analysis for representation of social interactions, taking into account the perception and a psychosocial profile of each individual. This model will facilitate the implementation of socially intelligent agents.
Dora-Luz Flores, Antonio Rodríguez-Díaz, Juan R. Castro, Carelia Gaxiola
Fuzzy Personality Model Based on Transactional Analysis and VSM for Socially Intelligent Agents and Robots
Abstract
This paper presents a model to add personality features to robots, the Transactional Analysis (TA) is used to define a psychological profile and the Viable Systems Model (VSM) is used to help explain the decisions made by the robot and how this impacts on its viability.
Carelia Gaxiola, Antonio Rodríguez-Díaz, Susan Jones, Manuel Castañón-Puga, Dora-Luz Flores

Robotics and Hardware Implementations

Frontmatter
Controlling Unstable Non-Minimum-Phase Systems with Fuzzy Logic: The Perturbed Case
Abstract
In this paper Fuzzy Logic Systems (FLS) for controlling non-minimumphase systems are proposed. A generalized Proportional Integral Derivative (PID) Fuzzy Logic Controller (FLC) for a benchmarking second order problemwith an unstable zero is presented. The same Fuzzy Rule Base (FRB) of the PID FLC is used in a PD FLC to regulate a plant consisting in a non-minimum-phase servomechanism with nonlinear backlash. Simulations demonstrate that the proposed FLC can be used to handle the non-minimum-phase systems.
Nohe R. Cazarez-Castro, Luis T. Aguilar, Oscar Castillo, Antonio Rodríguez-Dŕaz
Genetic Optimization for the Design of Walking Patterns of a Biped Robot
Abstract
This paper presents an Genetic Algorithm(GA) for the design of walking patterns of a 3-DOF biped robot formulated as a system with impulse effects. The GA optimizes the coefficients of a polynomialwhich represents the desired behavior of the walking which is included into the dynamics of the biped robot to obtain periodic motions while fulfills a minimal energy consumption over a complete walking cycle assumed as single support and instantaneous double support phases. Optimization results are presented showing walking patterns with low energy consumption and periodic motions.
Selene L. Cardenas-Maciel, Oscar Castillo, Luis T. Aguilar, Antonio Rodríguez-Díaz
Design and Simulation of the Type-2 Fuzzification Stage: Using Active Membership Functions
Abstract
This paper describes the design and implementation of the fuzzification stage for type-1 and type-2 fuzzy inference systems (FIS). A versatile method to calculate the membership values was used, it handles real numbers using decimal floating point binary encoding to calculate the slopes of triangular and trapezoidal membership functions. The designs were developed using VHDL code for FPGA implementation. The type-1 implementation is shown to give the basis of the type-2 implementation, which is based on the average method that consists in substituting an interval type-2 FIS by two type-1 FISs to cope with uncertainty. The functionality of the designs were evaluated by the analysis of the control surface plots of a speed controller for a DC motor. The plots were obtained from Simulink models that includes the VHDL designs developed in the Xilinx ISE. They were imported to the Simulink environment through the Xilinx System Generator.
Oscar Montiel, Roberto Sepúlveda, Yazmín Maldonado, Oscar Castillo
Methodology to Test and Validate a VHDL Inference Engine of a Type-2 FIS, through the Xilinx System Generator
Abstract
In this paper an improved high performance type-1 inference engine (IE) is proposed that can be applied with no modifications to the implementation of a type-2 FIS using the average method. The performance of the type-2 FIS will not be diminish for the use of this stage since it is achieved in parallel. The proposals are focused to be implemented into an FPGA. Simulink models to test the type-1 and type-2 inference engines are presented. The type-2 IE was tested in a speed controller for a DC motor.
Roberto Sepúlveda, Oscar Montiel, José Olivas, Oscar Castillo
Modeling and Simulation of the Defuzzification Stage of a Type-2 Fuzzy Controller Using the Xilinx System Generator and Simulink
Abstract
This paper is focused on the study, analysis and development of code for the defuzzification stage of type-2 fuzzy systems, through the average of two type-1 fuzzy systems. This proposal is based on the average method for systems where the type-2 membership functions of the inputs and output, have no uncertainty in the mean or center. The codification is done using the hardware description language VHDL, and it was exported to Simulink through the Xilinx System Generator (XSG). Comparative tests were conducted between the type-2 fuzzy systems for different number of bits and noise levels.
Roberto Sepúlveda, Oscar Montiel, Gabriel Lizárraga, Oscar Castillo
Backmatter
Metadata
Title
Evolutionary Design of Intelligent Systems in Modeling, Simulation and Control
Editors
Oscar Castillo
Witold Pedrycz
Janusz Kacprzyk
Copyright Year
2009
Publisher
Springer Berlin Heidelberg
Electronic ISBN
978-3-642-04514-1
Print ISBN
978-3-642-04513-4
DOI
https://doi.org/10.1007/978-3-642-04514-1

Premium Partners

    Image Credits