|
|
|
|
BOOKS/LIBROS |
||
|
Mining and Control of Network Traffic by Computational Intelligence F. Montesino Pouzols, D. R. Lopez, A. Barriga Series: Studies in Computational Intelligence, Vol. 342, Springer, 2011.
As other complex systems in social and natural sciences as well as in engineering,
the Internet is difficult to understand from a technical point of view. The structure
and behavior of packet switched networks is hard to model in a way comparable
to many natural and artificial systems. Nonetheless, the Internet is an outstanding
and challenging case due to its incredibly fast development and the inherent lack of
measurement and monitoring mechanisms in its core conception. In short, packet
switched networks defy analytical modeling. [DOI]
|
||
|
Fuzzy Logic-Based Algorithms for Video De-Interlacing P. Brox, I. Baturone, S. Sánchez-Solano Series: Studies in Fuzziness and Soft Computing, Vol. 246, Springer, 2010.
The 'Fuzzy Logic' research group of the Microelectronics Institute of Seville is
composed of researchers who have been doing research on fuzzy logic since the
beginning of the 1990s. Mainly, this research has been focused on the microelectronic
design of fuzzy logic-based systems using implementation techniques which
range from ASICs to FPGAs and DSPs. Another active line was the development
of a CAD environment, named Xfuzzy, to ease such design. Several versions of
Xfuzzy have been and are being currently developed by the group. The addressed
applications had basically belonged to the control field domain. In this sense, several
problems without a linear control solution had been studied thoroughly. Some
examples are the navigation control of an autonomous mobile robot and the level
control of a dosage system. [DOI]
|
||
|
Microelectronic Design of Fuzzy Logic-Based Systems I. Baturone, A. Barriga, S. Sánchez-Solano, C. J. Jiménez-Fernández, D.R. López CRC Press, March 2000. Fuzzy logic has virtually exploded over the landscape of emerging technologies, becoming an integral part of myriad applications and a standard tool for engineers. Until recently, most of the attention and applications have centered on fuzzy systems implemented in software. But these systems are limited. Problems that require real-time operation, low area, or low power consumption demand hardware designed to the fuzzy paradigm - and engineers with the background and skills to design it. Microelectronic Design of Fuzzy Logic-Based Systems offers low-cost answers to issues that software cannot resolve. From the theoretical, architectural, and technological foundation to design tools and applications, it serves as your guide to effective hardware realizations of fuzzy logic. Ø Review fuzzy logic theory and the basic issues of fuzzy sets, operators, and inference mechanisms Ø Explore the trade-offs between efficient theoretical behavior and practical hardware realizations Ø Discover the properties of the possible microelectronic realizations of fuzzy systems - conventional processors, fuzzy coprocessors, and fuzzy chips Ø Investigate the design of fuzzy chips that implement the whole fuzzy inference method into silicon Ø Analyze analog, digital, and mixed-signal techniques Ø Reduce your design effort for fuzzy systems with CAD tools - learn the requirements they should meet and survey current environments. Ø Put it all together - see examples and case studies illustrating how all of this is used to solve particular problems related to control and neuro-fuzzy applications
|
||
|
PAPERS/PUBLICACIONES |
||
|
CAD Tools for Fuzzy Logic Design HW/SW Codesign of Fuzzy Controllers Fuzzy Logic Applications for Mobile Robots Fuzzy Image Processing Fuzzy Systems for Traffic Control in Internet Digital Implementation of Fuzzy Controllers Mixed-Signal Implementation of Fuzzy Controllers |
||
|
CAD Tools for Fuzzy Logic Design Herramientas de CAD para Lógica Difusa |
||
|
M. Brox, S. Sánchez-Solano, E. del Toro, P. Brox, F. J. Moreno-Velo This paper describes two computer aided design (CAD) tools for automatic synthesis of fuzzy logic-based inference systems. The tools share a common architecture for efficient hardware implementation of fuzzy modules, but are based on two different design strategies. One of them is focused on the generation of standard VHDL code, which can be later implemented on a reconfigurable device (FPGA) or as an application specific integrated circuit (ASIC). The other one uses the Matlab/Simulink environment and tools for development of digital signal processing (DSP) systems on Xilinx's FPGAs. Both tools are included in the last version of $Xfuzzy$, a specific environment for designing complex fuzzy systems, and they provide interfaces to commercial VHDL synthesis and verification tools, as well as to conventional FPGA development environments. As demonstrated by the included design example, the proposed development strategies speed up the stages of description, synthesis, and functional verification of embedded fuzzy inference systems. [DOI]
F. J. Moreno-Velo, A. Barriga, S. Sánchez-Solano, I. Baturone This paper presents a new modeling language for fuzzy systems called XFSML. It is an XML-based language and it is proposed as a starting point for the definition of a standard modeling language in the fuzzy community. The main features of the language are its high expressiveness and its independence from specific platforms, tools or programming languages. [DOI]
M. Brox, L. L. Delgado, S. Sánchez-Solano This paper presents a design technique that allows the automatic synthesis of fuzzy inference systems and accelerates the exploration of the design space of these systems. It is based on generic VHDL code generation which can be implemented on a programmable device (FPGA) or an application specific integrated circuit (ASIC). The set of CAD tools supporting this technique includes a specific environment for designing fuzzy systems, in combination with commercial VHDL simulation and synthesis tools. As demonstrated by the analyzed design examples, the described development strategy speeds up the stages of description, synthesis, and functional verification of fuzzy inference systems. [DOI]
P. Martín, F. J. Moreno-Velo Most of the algorithms for extracting fuzzy classification rules generate conjunctive antecedents that use all the attributes of the system. Using this kind of antecedents, the number of rules grows exponentially in terms of the number of attributes of the system. This paper presents a new algorithm, FuzzyCN2, for extracting conjunctive fuzzy classification rules. This algorithm is a fuzzy version of the well known CN2 algorithm and produces an ordered list of fuzzy rules. FuzzyCN2 generates antecedents that may not include all the attributes of the system. These antecedents may cover a wide number of instances and, so, the number of extracted rules is smaller. The algorithm introduces the use of linguistic hedges as part of the selectors, thus producing more compact rules and reducing the number of generated rules. [DOI]
S. Sánchez-Solano, M. Brox, I. Baturone, A. Barriga Esta comunicación presenta una nueva técnica de implementación de sistemas difusos que está basada en el uso de una librería de módulos específicos, denominada XfuzzyLib, y cuyo flujo de diseño combina las herramientas de modelado y simulación del entorno Matlab con las de síntesis e implementación de FPGAs de Xilinx. La estrategia propuesta, que constituye la base de una nueva herramienta de síntesis hardware del entorno Xfuzzy, aprovecha las ventajas de flexibilidad y facilidad de configuración que brindan las diferentes herramientas de Matlab y Xilinx, permitiendo acelerar considerablemente las etapas de descripción, síntesis y verificación funcional de los sistemas bajo desarrollo.
F. J. Moreno-Velo, I. Baturone, S Sánchez-Solano, A. Barriga El entorno Xfuzzy 3 está formado por un amplio conjunto de herramientas dedicadas a dar soporte a las diferentes etapas del desarrollo de sistemas difusos. Entre estas herramientas se encuentra Xfdm, dedicada a la extracción de conocimiento difuso a partir de conjuntos de datos. Este trabajo presenta las últimas modificaciones realizadas en esta herramienta, que consisten en la integración de los algoritmos de clasificación difusos FuzzyID3 y FuzzConRI.
S. Sánchez-Solano, M. Brox, A. Cabrera En esta comunicación se describe una nueva técnica de implementación de sistemas difusos, basada en el uso de una librería de módulos específicos, cuyo flujo de diseño combina las herramientas de modelado y simulación del entorno Matlab y las herramientas de síntesis e implementación de FPGAs de Xilinx. La estrategia propuesta aprovecha las ventajas de flexibilidad y facilidad de configuración que brindan las diferentes herramientas de Matlab y Xilinx, permitiendo acelerar considerablemente las etapas de descripción, síntesis y verificación funcional de los sistemas difusos bajo desarrollo.
I. Baturone, F. J. Moreno-Velo, S. Sánchez-Solano, A. Barriga, P. Brox, A. Gersnoviez, M. Brox From 1992, Xfuzzy environment has been improving to ease the design of fuzzy systems. The current version, Xfuzzy 3, which is entirely programmed in Java, includes a wide set of new featured tools that allow automating the whole design process of a fuzzy logic based system: from its description (in the XFL3 language) to its synthesis in C, C++ or Java (to be included in software projects) or in VHDL (for hardware projects). The new features of the current version have been exploited in different application areas such as autonomous robot navigation and image processing.
F. J. Moreno-Velo, I. Baturone, A. Barriga, S. Sánchez-Solano Tuning a fuzzy system to meet a given set of requirements is usually a difficult task that involves many parameters. Since doing it manually is often cumbersome, several CAD tools have been reported to automate this process. The tool we have developed, xfsl, tries to reduce the limitations of other tools. In this sense, it includes a wide set of supervised learning algorithms and is able to cope with complex fuzzy systems. In particular, xfsl is able to adjust hierarchical fuzzy systems; systems that employ fuzzy functions defined freely by the user, like membership or connective functions, defuzzification methods, or even linguistic hedges; and fuzzy systems with continuous outputs (such as fuzzy controllers) as well as categorical outputs (such as fuzzy classifiers). Several examples included in this paper illustrate all these issues. Another relevant advantage is that xfsl is integrated into the fuzzy system development environment Xfuzzy 3, and, hence, it can be easily employed within the design flow of a fuzzy system.
A. Gersnoviez, I. Baturone, F. J. Moreno-Velo Este artículo presenta una técnica basada en la lógica difusa para extraer bases de reglas a partir de datos numéricos. Permite obtener bases de reglas interpretables lingüísticamente a la vez que simples en cuanto a número de reglas, sencillez en las partes antecedentes y consecuentes y facilidad de implementación hardware/software. Los pasos más significativos de esta técnica son los siguientes: (1) extracción de la base de reglas empleando particiones granulares de las variables del problema, (2) ajuste de las funciones de pertenencia para las variables de salida y posterior simplificación, (3) simplificación tabular de la base de reglas y (4) simplificación de las funciones de pertenencia para las variables de entrada. La técnica puede aplicarse de forma automática mediante las herramientas de CAD integradas en el entorno Xfuzzy 3. Se incluye un ejemplo de aplicación en robótica móvil para ilustrar las ventajas de la técnica propuesta. [Pdf]
A. Barriga, S. Sánchez-Solano, I. Baturone, D. R. López, F. J. Moreno-Velo,
P. Brox, F. Montesino, N. M. Hussein, M. Brox, A. Gersnoviez, The characteristics of the new version of the fuzzy systems development environment Xfuzzy is presented. The environment covers the aspects related to the specification, verification, adjustment and implementation of fuzzy systems. It is an open environment (in the sense that the user can define many functional and structural aspects) and a free distribution tool that allows proving new formalisms and helps the definition and implementation of complex systems. [Pdf]
F. J. Moreno Velo, I. Baturone, S. Sánchez Solano, A. Barriga, Este trabajo presenta un estudio sobre las diferentes formas de definir conceptos lingüísticos difusos, ya sean funciones de pertenencia libres o familias de funciones de pertenencia, y describe la capacidad del lenguaje XFL3 (el lenguaje de especificación formal de Xfuzzy 3) tanto para utilizar ambas formas como para extender el conjunto de funciones y familias disponibles. El trabajo incluye un ejemplo de aplicación en el que se pone de manifiesto las ventajas e inconvenientes de cada opción. [Pdf]
F. J. Moreno-Velo, I. Baturone, S. Sánchez-Solano , A. Barriga This paper presents a study of the different kinds of membership function (MF) definitions, regarding free MFs and families of MFs, and describes the capabilities of XFL3 (the formal specification language defined by Xfuzzy 3) to manage them. This includes not only the possibility of using them in a system design, but also the capability for extending the available functions with new user defined membership functions and families. An application example has been included in order to discuss on the suitable parametric definition of the functions.
F. J. Moreno-Velo, I. Baturone, R. Senhadji, S. Sánchez-Solano Tuning a fuzzy system to meet a given set of input/output patterns is usually a difficult task that involves many parameters. This paper presents an study of different approaches that can be applied to perform this tuning process automatically, and describes a CAD tool, named xfsl, which allows applying a wide set of these approaches: (a) a large number of supervised learning algorithms; (b) different processes to simplify the learned system; (c) tuning only specific parameters of the system; (d) the ability to tune hierarchical fuzzy systems, systems with continuous output (like fuzzy controller) as well as with categorical output (like fuzzy classifiers), and even systems that employ user-defined fuzzy functions; and, finally, (e) the ability to employ this tuning within the design flow of a fuzzy system, because xfsl is integrated into the fuzzy system development environment Xfuzzy 3.0.
R. Senhadji, S. Sánchez-Solano, A. Barriga, I. Baturone, F. J. Moreno-Velo This contribution presents a new algorithm (NORFREA) to select fuzzy rules from a grid partition of the input domain. Besides using an efficiency measure for the rules, this algorithm employs an heuristic technique to reduce the influence of the initial grid structure. Different benchmarks of classification problems are included to illustrate the advantages of this algorithm.
F. J. Moreno Velo, I. Baturone, S. Sánchez-Solano, A. Barriga, R. Senhadj Este trabajo presenta el estado actual de la nueva versión del entorno Xfuzzy de desarrollo de sistemas difusos. El entorno está compuesto por un amplio conjunto de herramientas que cubren las diferentes etapas del diseño de estos sistemas. Estas herramientas comparten una descripción común en un lenguaje de descripción formal llamado XFL3. Las principales novedades de esta versión son el uso de este nuevo lenguaje, más potente y flexible que en versiones anteriores, la inclusión de nuevas herramientas, y la posibilidad de ejecutar el entorno en cualquier sistema operativo, ya que ha sido programado totalmente en lenguaje Java.
R. Senhadji, I. Baturone, A. Barriga, S. Sánchez-Solano,F.J. Moreno Velo En esta contribución se presenta un nuevo algoritmo (NORFREA) para seleccionar reglas difusas desde una estructura de grid. Además de la correspondiente medida de eficiencia de las reglas, el algoritmo utiliza una heurística para intentar mitigar la influencia de la partición inicial en la eficiencia del sistema. Diferentes benchmarks para problemas de clasificación han sido utilizados para probar la eficacia del algoritmo propuesto.
A. Cabrera, R. Senhadji, S. Sánchez-Solano, A. Barriga, C. J. Jiménez, O. Llanes This paper describes the development of different kinds of level controllers based on fuzzy logic. Designs and implementations were carried out using tools from xfuzzy, a development environment that eases the different stages in the design of fuzzy inference systems. Special emphasis has been put in the on-line verification of the controller over a physical plant by means of xflab tool. Different approaches of the knowledge base were tested using xflab. Then, the results were analyzed and selected those that gave the better performance. Controllers implementations with different number of bit for input / output resolution were also carried out and analyzed. The results provide a base for the incoming development of a hardware fuzzy logic controller by means of specific hardware or by an embedded codesign system. [Pdf]
F. J. Moreno Velo, S. Sánchez Solano, A. Barriga, I. Baturone, D.R. López This work presents de main features of XFL3, a language for fuzzy system specification, which has been defined as the common description language for the tools forming the Xfuzzy 3.0 development environment. Its main advantages are its capability to admit user-defined membership functions, parametric operators, and linguistic hedges. A brief summary of the tools included in Xfuzzy 3.0 and an example illustrating the use of XFL3 are also included. [Pdf]
A. Bermúdez, A. Barriga, I. Baturone, S. Sánchez-Solano FLEB is an electronic book which attempts to introduce the basic mathematical foundations and applications of fuzzy logic through a software environment which includes images, hypertext, sensitive elements, animations and interactive demos. It also allows executing Xfuzzy, a development tool which eases the description, verification, and synthesis of fuzzy logic-based systems. FLEB, like a usual book, is structured into chapters with pages through which the reader can navigate comfortably. In addition, the information provided can be accessed in a non sequential way thanks to the hypertext and sensitive elements that interconnect linked pages. This capability of non sequential reading together with the exploitation of multimedia software make FLEB a good tool to pedagogically show and explain the basis of fuzzy logic theory and applications. [Pdf]
F. J. Moreno-Velo, I. Baturone, S. Sánchez-Solano, A. Barriga This paper presents Xfsl, a tool for the automatic tuning of fuzzy systems using supervised learning algorithms. The tool provides a wide set of learning algorithms, which can be used to tune complex systems. An important issue is that Xfsl is integrated into the fuzzy system development environment Xfuzzy 3.0, and hence, it can be easily employed within the design flow of a fuzzy system. [Pdf]
F. J. Moreno-Velo, I. Baturone, S. Sánchez-Solano, A. Barriga This paper presents the new version of Xfuzzy, Xfuzzy 3.0, which is a development environment for fuzzy-inference-based systems. It is composed by many tools that cover the different stages of the fuzzy system design process, from their initial description to the final implementation. Its main features are the capability for developing complex systems, the flexibility of allowing the user to extend the set of available functions and the possibility of been executed on any platform with JRE (Java Runtime Environment) installed. [Pdf]
F. J. Moreno-Velo, S. Sánchez-Solano, A. Barriga, I. Baturone, D. R. López This paper presents the main features of XFL3, a new language for fuzzy system specification, which has been defined as the starting point for the 3.0 version of our fuzzy system design environment, Xfuzzy [1]. Its main advantages with respect to its precursor, XFL [2], are its capability to admit user-defined membership functions, parametric operators, and linguistic hedges. Taking this language as the basis, different fuzzy system development tools are being implementing, which are also summarized briefly. [Pdf]
A. Bermúdez, A. Barriga, I. Baturone, S. Sánchez-Solano FLEB son las siglas de Fuzzy Logic E-Book, esto es, Libro Electrónico sobre Lógica Difusa. Pretende ser una introducción a los fundamentos y aplicaciones de la lógica difusa a través de un entorno informático que integra imágenes, hipertexto, elementos sensibles, animaciones y demos interactivas. Permite además la ejecución de Xfuzzy, una herramienta de desarrollo que facilita la descripción, verificación y síntesis de sistemas basados en lógica difusa. El entorno facilita la navegación a lo ancho de todo el contenido, que se encuentra estructurado en capítulos divididos en páginas (de ahí la denominación de libro electrónico). Asimismo, el hipertexto y los elementos sensibles conectan unas páginas con otras permitiendo un acceso no secuencial a la información. Esta posibilidad de moverse directamente entre páginas no necesariamente consecutivas pero relacionadas entre sí (páginas enlazadas) constituye la principal característica del libro electrónico, junto con las facilidades ofrecidas por la informática multimedia para la exposición y comprensión del tema tratado, en este caso la lógica difusa.
F. J. Moreno Velo, S. Sánchez Solano, A. Barriga, I. Baturone, D.R. López En esta comunicación presentamos las principales características de XFL3, un nuevo lenguaje de especificación de sistemas difusos que ha sido definido como punto de partida para la versión 3.0 del entorno de desarrollo de sistemas difusos Xfuzzy. Sus principales ventajas respecto a su predecesor XFL son la capacidad de incorporar funciones de pertenencia definidas por el usuario, la posibilidad de utilización de operadores paramétricos y la incorporación de modificadores lingüísticos. Tomando como base este lenguaje se vienen implementando diferentes herramientas de desarrollo de las que se presenta una breve descripción.
R. Senhadji, I. Álvarez, A. Barriga, S. Sánchez Solano En esta comunicación se describe la realización de un controlador de nivel para sistemas de dosificación basado en lógica difusa. El diseño y la implementación se han llevado a cabo con ayuda de las diferentes herramientas del entorno de desarrollo Xfuzzy. Se incluyen asimismo resultados experimentales medidos sobre una planta prototipo.
R. Senhadji, S. Sánchez-Solano, D.R. López, A. Barriga This paper shows an on-line verification tool for fuzzy controllers. On-line verification means the simulation of the software model of the controller using the real plant in a closed loop system. The tool is part of a fuzzy logic system development environment that permits the inclusion of the tool into a design flow methodology for fuzzy systems. [Pdf]
I. Baturone, S. Sánchez-Solano, A. Barriga, C. J. Jiménez, D. R. López A development environment for fuzzy logic based inference systems is presented in this paper. CAD tools included within the environment simplify the tasks of specification, verification and synthesis of fuzzy systems. System specification is carried out with the help of a high-level description language. This specification is the input for all the verification and synthesis facilities provided by the environment. Verification tools allow to simulate the system behavior and to adjust the parameters defining the knowledge base. Finally, synthesis tools provide software and hardware implementations of fuzzy systems.
D. R. López, C. J. Jiménez, I. Baturone, A. Barriga, S. Sánchez-Solano Xfuzzy is a CAD tool that eases the development of fuzzy systems from their conception to their final implementation. It is composed of a set of modules and programs that share a common specification language and cover the different stages of the design process. Modules for describing, verifying and tuning the behavior of the system are integrated within the environment. In addition to these features, common to other fuzzy design tools, a relevant characteristic of Xfuzzy is that it includes several synthesis facilities for implementing the system on either software or hardware.
E. Lago, C. J. Jiménez, D. R. López, S. Sánchez-Solano, A. Barriga A tool for the synthesis of fuzzy controllers is presented in this paper. This tool takes as input the behavioral specification of a controller and generates its VHDL description according to a target architecture. The VHDL code can be synthesized by means of two implementation methodologies, ASIC and FPGA. The main advantages of using this approach are rapid prototyping, and the use of well-known commercial design environments like Synopsys, Mentor Graphics, or Cadence.
D. López, F. J. Moreno, S. Sánchez-Solano, A. Barriga Xfuzzy 2.0 constituye un entorno para la descripción, verificación y síntesis de sistemas basados en lógica difusa. Entre sus características cabe destacar la utilización de un lenguaje de especificación potente y flexible para describir la estructura de un sistema y definir las operaciones difusas involucradas. El entorno Xfuzzy incluye una serie de herramientas gráficas que facilitan la descripción de un sistema difuso, así como utilidades para verificar y refinar su comportamiento. También incorpora módulos de síntesis de sistemas difusos, que permiten obtener realizaciones software y hardware de los mismos.
C. J. Jiménez, D. López, E. Lago, M. A. Hinojosa, A. Barriga, S. Sánchez-Solano En esta comunicación se describen las herramientas de síntesis de controladores difusos incluidas en Xfuzzy 2.0, un entorno de descripción, verificación y síntesis de sistemas basados en lógica difusa. Estas herramientas permiten dos estrategias de diseño (tabla look-up o arquitectura específica del controlador) y dos técnicas de implementación (FPGAs o ASICs). Se evalúan las herramientas mediante ejemplos de diseños que nos permiten extraer conclusiones sobre la aplicabilidad de cada solución.
F. J. Moreno, D. López, A. Barriga, S. Sánchez-Solano This paper introduces the main features of XFBPA, a tool that provides automatic learning capabilities for fuzzy systems, integrated into the fuzzy development environment Xfuzzy. Its main achievement is the support for the learning of complex systems, preserving the flexibility in system description provided by Xfuzzy. [Pdf]
D. López, F. J. Moreno, A. Barriga, S. Sánchez-Solano This paper presents the main characteristics of XFL, a language for fuzzy logic based systems. Its more relevant features are the ability to define potentially complex systems and the independence from the particular techniques employed for implementing fuzzy operations. This language has been used as the base for several hardware- and software-oriented development tools, which are also briefly introduced.
A. Barriga, S. Sánchez-Solano, C. J. Jiménez, D. Gálan, D. R. López, This paper describes a design environment for the hardware realizations of fuzzy controllers which includes a set of CAD tools to ease the description, verification and synthesis of this kind of systems. Special emphasis is focused on the use of a standard hardware description language (VHDL) and compatibility with other integrated circuits design tools. [Pdf]
D. R. López, S. Sánchez-Solano, A. Barriga En esta comunicación describimos las principales características de XFL, un lenguaje para la especificación de sistemas basados en lógica difusa que permite realizar definiciones independientemente de las técnicas particulares empleadas para las operaciones difusas. XFL permite también el empleo de bases de reglas jerárquicas para la definición de sistemas arbitrariamente complejos. Este lenguaje ha permitido el desarrollo de varias herramientas de desarrollo (hardware y software), de las que se presenta una breve descripción.
D. Galán, C. J. Jiménez, A. Barriga, S. Sánchez-Solano The particular architecture of fuzzy systems has led to the introduction of specific simulators on the market, usually isolated from design environments. This article presents a VHDL package that allows high level descriptions and simulations of fuzzy controllers. The importance of this package to fuzzy hardware design lies not only in its portability to any VHDL simulator, but also in that it allows verification of the simulation results of a particular system in a unique and standard simulation environment, from algorithm description level to RTL level (synthesizable) or logic gates. Finally, an example is included to demonstrate the package functionality.
C. J. Jiménez, D. Gálan, A. Barriga, S. Sánchez-Solano Se describe un entorno de diseño para la realización hardware de sistemas de control basados en lógica difusa. Dicho entorno combina una serie de herramientas de CAD que facilitan la definición de las especificaciones del controlador, la descripción de su arquitectura y la síntesis automática del circuito. Entre sus principales características cabe destacar la utilización de un lenguaje de descripción de hardware estándar (VHDL) y su compatibilidad con otras herramientas de diseño de circuitos integrados.
| ||
|
HW/SW Codesign of Fuzzy Controllers Codiseño HW/SW de Controladores Difusos |
||
|
S. Sánchez-Solano, E. del Toro, M. Brox, P. Brox, I. Baturone The complexity reached by current applications of industrial control systems has motivated the development of new computational paradigms, as well as the employment of hybrid implementation techniques that combine hardware and software components to fulfill systems requirements. On the other hand, continuous improvements in field programmable devices make today possible the implementation of complex control systems on reconfigurable hardware, although they are limited by the lack of specific design tools and methodologies to facilitate the development of new products. This paper describes a model-based design approach for the synthesis of embedded fuzzy controllers on FPGAs. Its main contributions are the proposal of a novel implementation technique, which allows accelerating the exploration of the design space of fuzzy inference modules, and the use of a design flow that eases their integration into complex control systems and the joint development of hardware and software components. This design flow is supported by specific tools for fuzzy systems development and standard FPGA synthesis and implementation tools, which use the modeling and simulation facilities provided by the Matlab environment. The development of a complex control system for parking an autonomous vehicle demonstrates the capabilities of the proposed procedure to dramatically speed up the stages of description, synthesis, and functional verification of embedded fuzzy controllers for industrial applications. [DOI]
S. Sánchez-Solano, E. del Toro, M. Brox, I. Baturone, A. Barriga This paper presents a design environment for the synthesis of embedded fuzzy controllers on FPGAs. It provides a novel implementation technique that allows accelerating the exploration of the design space of fuzzy control modules, as well as a codesign flow that eases their integration into complex control systems and the joint development of hardware and software components. The set of CAD tools supporting this environment includes specific fuzzy logic design tools provided by Xfuzzy, FPGA synthesis and implementation tools from Xilinx, and modeling and simulation facilities from Matlab. As demonstrated by the analyzed design examples, the described development strategy takes advantage of flexibility and ease of configuration offered by the different tools to dramatically speed up the stages of description, synthesis, and functional verification of embedded fuzzy control systems. [DOI]
S. Sánchez-Solano, M. Brox, E. del Toro, A. J. Cabreraa This paper describes a hardware/software codesign strategy for fuzzy control systems implementation using FPGAs. The main contribution of the paper consists of a methodology for joint development of hardware and software components intended for rapid and verifiable design of a fuzzy control system. The design flow combines specific tools for fuzzy inference systems included in the XFuzzy environment, simulation and modelling tools from Matlab and FPGA synthesis, and implementation tools provided by Xilinx. The advantages of this proposal are described in section 4 as it is used for the control system development of an autonomous vehicle.
S. Sánchez-Solano, A. Cabrera, I. Baturone, A. Barriga, F. J. Moreno-Velo, P. Brox En esta comunicación se describe la realización mediante técnicas de codiseño hardware/software de un sistema de control difuso para aplicaciones de navegación de robots móviles. La implementación física del controlador se lleva a cabo por medio de una plataforma reconfigurable basada en FPGAs. Dicha plataforma, combina un procesador de propósito general con hardware específico para la implementación de módulos de inferencia difusos. Tanto el sistema de procesado como los módulos de inferencia son configurables, disponiéndose de herramientas de CAD que facilitan el desarrollo de los controladores. [Pdf]
A. Cabrera, S. Sánchez-Solano, P. Brox, A. Barriga, R. Senhadji Fuzzy inference techniques are an attractive and well-established approach for solving control problems. This is mainly due to their inherent ability to obtain robust, low-cost controllers from the intuitive (and usually ambiguous or incomplete) linguistic rules used by human operators when describing the control process. This paper focuses on the hardware/software codesign of configurable fuzzy control systems. Two prototype systems implemented on general-purpose development boards are presented. In both of them, hardware components are based on specific and configurable fuzzy inference architecture whereas software tasks are supported by a microcontroller. The first prototype uses an off-the-shelf microcontroller and a low-complexity Xilinx XC4005XL field programmable gate array (FPGA). The second one is implemented as a system on programmable chip (SoPC), integrating the microcontroller together with the fuzzy hardware architecture and its interface circuits into a Xilinx Spartan2E200 FPGA. [DOI]
A. Cabrera, S. Sánchez-Solano, I. Baturone, F. J. Moreno-Velo, A. Barriga En el presente trabajo se describe la realización de una plataforma basada en FPGAs de Xilinx que permite el desarrollo de sistemas de control difusos. La plataforma, que incorpora un procesador MicroBlaze, facilita la implementación de controladores difusos totalmente software o basados en codiseño hardware/software en donde el módulo de inferencia se implementa mediante hardware específico. Tanto el sistema de procesado basado en MicroBlaze como el hardware del módulo de inferencia son configurables, disponiéndose de herramientas de CAD para el desarrollo integral del controlador. [Pdf]
A. Cabrera, S. Sánchez-Solano, R. Senhadji, A. Barriga, C. J. Jimémez En este trabajo se describe una metodología de codiseño HW/SW para el desarrollo de controladores difusos soportados sobre una plataforma compuesta por un microcontrolador y una FPGA. Se exponen las diferentes fases de la metodología, así como las herramientas de CAD utilizadas, con énfasis en el entorno de desarrollo de sistemas Xfuzzy. Por último, se incluye una aplicación práctica de la metodología descrita al desarrollo de un controlador difuso para un sistema de dosificación. Los resultados obtenidos permiten desarrollar controladores embebidos en donde se incluya el controlador en el propio FPGA.
A. Cabrera, S. Sánchez-Solano, R. Senhadj, A. Barriga, C. J. Jimémez, I. Baturone En esta comunicación se describe la realización mediante codiseño Hardware/Software de un controlador difuso para un sistema de dosificación de líquidos. La implementación final se ha realizado utilizando una placa comercial de desarrollo de FPGAs compuesta por un microcontrolador 8031 y una FPGA XC4005XL. Para el diseño del controlador difuso se ha utilizado el entorno de desarrollo de sistemas difusos Xfuzzy, de libre distribución, mientras que la síntesis hardware del sistema se ha llevado a cabo mediante versiones educativas de herramientas comerciales.
S. Sánchez-Solano, R. Senhadji, A. Cabrera, I. Baturone, C. J. Jiménez, A. Barriga This paper describes a design methodology for fuzzy logic-based control systems. The methodology employs hardware/software codesign techniques according to an 'a priori' partition of the tasks assigned to the selected components. This feature makes it possible to tackle the control system prototyping as one of the design stages. In our case, the platform considered for prototyping has been a development board containing a standard microcontroller and an FPGA. Experimental results from an actual control application validate the efficiency of this methodology.
A. Cabrera, S. Sánchez-Solano, R. Senhadji, A. Barriga, C. J. Jiménez This paper describes a HW/SW codesign methodology for the implementation of fuzzy controllers on a platform composed by a general-purpose microcontroller and specific processing elements implemented on FPGAs or ASICs. The different phases of the methodology, as well as the CAD tools used in each design stage, are presented, with emphasis on the fuzzy system development environment Xfuzzy. Also included is a practical application of the described methodology for the development of a fuzzy controller for a dosage system.
A. Cabrera, S. Sánchez-Solano, R. Senhadji, A. Barriga, C. J. Jiménez En esta comunicación se describe una metodología de codiseño HW/SW para la implementación de controladores difusos utilizando microcontroladores de propósito general y elementos específicos de procesado construidos mediante FPGAs o ASICs. Se exponen las diferentes fases de la metodología, así como las herramientas de CAD utilizadas para el desarrollo de las mismas, con énfasis en el entorno de desarrollo de sistemas difusos Xfuzzy. Por último, se incluye una aplicación práctica de la metodología descrita al diseño de un controlador difuso para un sistema de dosificación.
| ||
|
Fuzzy Logic Applications for Mobile Robots Aplicaciones de Lógica Difusa en Robótica Móvil |
||
|
I. Baturone, F. J. Moreno-Velo, V. Blanco, J. Ferruz Fuzzy controllers are used in many applications because of their rapid design by translating heuristic knowledge, robustness against perturbations, and smoothness in the control action. However, they require parallel processing and special operators (such as fuzzification or defuzzification) which are not available in standard digital signal processors (DSPs), thus complicating their direct implementation. This paper describes an efficient design methodology that allows starting with any kind of fuzzy controller and subsequently transforming it until a system suitable for easy DSP implementation is obtained. Such methodology is greatly aided by the design environment Xfuzzy 3. The parking problem of an autonomous robot is described to illustrate the steps of this methodology. Real experiments with the autonomous robot ROMEO 4R demonstrate efficiency of the designed fuzzy controller embedded into a stand-alone card based on a fixed-point DSP from Texas Instruments. [DOI]
S. Sánchez-Solano, A. J. Cabrera, I. Baturone, F. J. Moreno-Velo, M. Brox Fuzzy logic-based inference techniques provide efficient solutions for control problems in classical and emerging applications. However, the lack of specific design tools and systematic approaches for hardware implementation of complex fuzzy controllers limits the applicability of these techniques in modern microelectronics products. This paper discusses a design strategy which eases the implementation of embedded fuzzy controllers as systems on programmable chips (SoPC). The development of the controllers is carried out by means of a reconfigurable platform based on field-programmable gate arrays (FPGAs). This platform combines specific hardware to implement fuzzy inference modules with a general-purpose processor, thus allowing the realization of hybrid hardware/software solutions. As happens to the components of the processing system, the specific fuzzy elements are conceived as configurable Intellectual Property (IP) modules in order to accelerate the controller design cycle. The design methodology and tool chain presented in this paper have been applied to the realization of a control system for solving navigation tasks of an autonomous vehicle.
M. Brox, S. Sánchez-Solano This paper describes the design of fuzzy logic-based controllers as IP modules compatible with the processing systems available in the current families from Xilinx. These modules can be used as standard peripherals connectable to the OPB bus for the processors available in FPGAs, such as MicroBlaze or Power PC. The use of inference techniques based on fuzzy logic allows the development of complex control systems from the linguistic description of the knowledge of an expert operator, without any need to employ mathematical models, and provides good robustness in the face of changing operational conditions.
M. Brox, A. Gersnoviez, S. Sánchez-Solano, I. Baturone En esta comunicación se describe un sistema de control difuso para aplicaciones de navegación de robots móviles autónomos en presencia de obstáculos fijos. Las herramientas de CAD del entorno Xfuzzy 3, desarrollado en el IMSE, han facilitado el diseño del controlador. En la comunicación se procede a la verificación del controlador diseñado operando en un lazo cerrado con el modelo del robot móvil autónomo eléctrico Romeo 4R, diseñado y construido en la Escuela Superior de Ingenieros de la Universidad de Sevilla. Las simulaciones realizadas demuestran la eficiencia del controlador desarrollado. [Pdf]
I. Baturone, F. J. Moreno-Velo, S. Sánchez-Solano, V. Blanco, J. Ferruz Fuzzy controllers are used in many applications because of their rapid design by translating heuristic knowledge, robustness against perturbations, and smoothness in the control action. However, they require parallel processing and special operators (such as fuzzification or defuzzification) which are not available at standard DSPs, thus making inefficient its direct implementation. This paper describes a design methodology which allows starting with any kind of fuzzy controller and subsequently transforming it until obtaining a system suitable for DSP implementation. Such methodology is aided by Xfuzzy 3, a design environment developed by some of the authors. The parking problem of an autonomous robot is described to illustrate the steps of this methodology. Experimental results show the efficiency of the designed fuzzy controller embedded into a stand-alone card based on a fixed-point DSP from Texas Instruments.
I. Baturone, F. J. Moreno-Velo, S. Sánchez-Solano, V. Blanco, J. Ferruz This paper describes how to design fuzzy controllers meeting the hardware constraints of digital signal processors (DSPs). Fuzzy controllers are used in many applications because of their rapid design by translating heuristic knowledge, robustness against perturbations, and smoothness in the control action. However, they require parallel processing and special operators (like fuzzification or defuzzification) which are not available at standard DSPs, thus making unefficient its direct implementation. The idea followed in this paper is to translate the fuzzy rule bases of a fuzzy controller into non fuzzy ones that can be implemented easily by using the relational and logical operators, the standard if-then conditional statements, and the addition and multiplication operators available at a DSP. This is done by using hierarchical structures and adequate membership functions, connective operators, and inference methods. The parking problem of an autonomous robot is described to illustrate this design process. Experimental results show the efficiency of the designed fuzzy controller embedded into a stand-alone card based on a fixed-point DSP from Texas Instruments. [Pdf]
A. Cabrera, S. Sánchez-Solano, A. Barriga, P. Brox, F. J. Moreno-Velo, I. Baturone En esta comunicación se describen los diferentes componentes físicos, lógicos y de procedimiento de una plataforma que permite implementar una estrategia de realización de controladores difusos empotrados, tanto híbridos hardware/software como puramente software. Dicha estrategia está basada en la utilización de módulos de Propiedad Intelectual para la realización del sistema de procesamiento y de una arquitectura hardware específica para la implementación de los módulos de inferencia difusos, de forma tal que facilita la realización de controladores difusos como sistemas empotrados sobre dispositivos programables. [Pdf]
A. Cabrera, S. Sánchez-Solano, I. Baturone, F. J. Moreno-velo, P. Brox, A. Barriga This paper describes the realization of embedded fuzzy control systems for planning the motion of autonomous mobile robots. The development of the controllers is carried out by means of a reconfigurable platform based on FPGAs. This platform combines a general-purpose processor with specific hardware to implement fuzzy inference modules, thus allowing the comparison between a fully software solution and others based on hybrid hardware/software techniques. Both the processing system and the inference modules are configurable using available CAD tools, which make the development of the controllers easier.
A. Cabrera, S. Sánchez-Solano, I. Baturone, A. Barriga, F. J. Moreno-Velo, P. Brox En este trabajo se describe la realización de controladores difusos empotrados sobre un dispositivo programable para aplicaciones de navegación de un robot móvil. El desarrollo de los controladores se lleva a cabo mediante una plataforma reconfigurable que permite realizaciones híbridas hardware/software así como puramente software de los mismos. Esta plataforma está basada en la utilización de módulos de propiedad intelectual para la realización de un sistema de procesamiento configurable así como de una arquitectura también configurable para la realización del módulo de inferencia, empotrándose todo el controlador sobre un FPGA con la ayuda de las correspondientes herramientas de CAD. [Pdf]
I. Baturone, F. J. Moreno-Velo, S. Sánchez-Solano, A. Ollero This paper describes the design and implementation of a fuzzy control system for a car-like autonomous vehicle. The problem addressed is the diagonal parking in a constrained space, a typical problem in motion control of nonholonomic robots. The architecture proposed for the fuzzy controller is a hierarchical scheme which combines seven modules working in series and in parallel. The rules of each module employ the adequate fuzzy operators for its task (making a decision or generating a smoothly varying control output), and they have been obtained from heuristic knowledge and numerical data (with geometric information) depending on the module requirements (some of them are constrained to provide paths of near-minimal lengths). The CAD tools of the environment Xfuzzy 3.0 (developed by some of the authors) have been employed to automate the different design stages: (a) translation of heuristic knowledge into fuzzy rules, (b) extraction of fuzzy rules from numerical data and their tuning to give paths of near-minimal lengths, (c) off-line verification of the control system behavior, and (d) its synthesis to be implemented in a true robot and be verified on line. Real experiments with the autonomous vehicle ROMEO 4R (designed and built at the "Escuela Superior de Ingenieros", University of Seville) demonstrate the efficiency of the described controller and of the methodology followed in its design.
F. J. Moreno-Velo, I. Baturone, F. J. Barrio, S. Sánchez-Solano, A. Barriga The growing use of fuzzy systems in application domains of increasing complexity is making necessary the employment of a design methodology which arranges and facilitates the flow of the different design tasks. This paper presents a design methodology based on a model suitable for describing complex fuzzy systems, and a set of CAD tools aimed at supporting the different design tasks of the proposed methodology. An example on how to design a complex hierarchical system is included to illustrate this methodology. [Pdf]
I. Baturone, F. J. Moreno-Velo, S. Sánchez-Solano, A. Ollero This paper presents methods and tools to design a fuzzy path planner for autonomous non-holonomic vehicles by means of supervised learning. The method combines heuristic knowledge and geometric considerations to obtain a continuous-curvature short path that can be executed efficiently by the path tracking controller of the mobile robot. Furthermore, the method minimizes the computer requirements to implement the fuzzy planner. The proposed design method can be easily carried out by means of the Xfuzzy 3.0 environment developed by some of the authors. The resulting planning strategies have been proven successfully in the Romeo 4R autonomous vehicle fully designed and built at the Escuela Superior de Ingenieros, University of Seville. [Pdf]
F. J. Moreno-Velo, I. Baturone, S. Sánchez-Solano, A. Barriga The crecient use of fuzzy systems in complex applications has motivated us to develop a new version of Xfuzzy, the design environment for fuzzy system created at the IMSE (Instituto de Microelectrónica de Sevilla). This new version, Xfuzzy 3.0, offers the advantages of being enterely programmed in Java, and allows designing hierarchical rule bases that can interchange fuzzy or non fuzzy values as well as employ user-defined fuzzy connectives, linguistic hedges, membership functions, and defuzzification methods. Xfuzzy 3.0 integrates tools that facilitate the description, tuning, verification, and synthesis of complex fuzzy systems. This is illustrated in this paper with the design of a fuzzy controller to solve a parking problem. [Pdf]
I. Baturone, F. J. Moreno-Velo, S. Sánchez-Solano, R. Martín de Agar, A. Ollero This paper describes the design and implementation of a fuzzy controller for parking autonomous mobile robots. The fuzzy controller is a hierarchical system that emulates what we would do as drivers. As a consequence, it firstly decides if the robot has to drive forward or backward (at high or small speed) or has to brake, and, secondly, the adequate steering angle. The rule bases of the designed controller translate our expert knowledge expressed linguistically and, hence, they include relational concepts and linguistic hedges. The controller has been easily designed with the aid of the Xfuzzy 3.0 environment (developed at the "Instituto de Microelectrónica de Sevilla"), which contains CAD tools to specify, adjust, verify, and synthesize a fuzzy system. The efficiency of the designed controller has been successfully proven with the ROMEO 4R autonomous vehicle designed and built at the "Escuela Superior de Ingenieros", University of Seville. The results obtained with the real vehicle are similar to the simulation results and Romeo 4R is able to perform successfully a parking maneuver even when the vehicle is initially close to the parking position, with a bad orientation, and having to maneuver autonomously for parking. [Pdf]
| ||
|
Fuzzy Image Processing Procesado de Imágenes Mediante Lógica Difusa |
||
|
P. Brox, I. Baturone, S. Sánchez-Solano This paper presents the application of soft computing techniques to video processing. Especially, the research work has been focused on the de-interlacing task. It is necessary whenever the transmission standard uses an interlaced format but the receiver requires a progressive scanning, as happens in consumer displays such as LCDs and plasma. A simple hierarchical solution that combines three simple fuzzy logic-based constituents (interpolators) is presented in this paper. Each interpolator is specialized in one of three key image features for de-interlacing: motion, edges, and possible repetition of picture areas. The resulting algorithm offers better results than others with less or similar computational cost. A very interesting result is that our algorithm is competitive with motion-compensated algorithms. [DOI]
J. Gutiérrez-Ríos, P. Brox, F. Fernández-Hernández, I. Baturone, S. Sánchez-Solano Interlacing techniques were introduced in the early analog TV transmission systems as an efficient mechanism capable of halving the video bandwidth. Currently, interlacing is also used by some modern digital TV transmission systems, however, there is a problem at the receiver side since the majority of modern display devices require a progressive scanning. De-interlacing algorithms convert an interlaced video signal into a progressive one by performing interpolation. To achieve good de-interlacing results, dynamical and local image features should be considered. The gradual adaptation of the de-interlacing technique as a function of the level of motion detected in each pixel is a powerful method that can be carried out by means of fuzzy inference. The starting point of our study is an algorithm that uses a fuzzy inference system to evaluate motion locally (FMA algorithm). Our approach is based on convolution techniques to process a fuzzy rulebase for motion-adaptive de-interlacing. Different strategies based on bi-dimensional convolution techniques are proposed. In particular, the algorithm called 'single convolution algorithm' introduces significant advantages: a more accurate measurement of the level of motion using a matrix of weights, and a unique fuzzification process after the global estimation, which reduces the computational cost. Different architectures for the hardware implementation of this algorithm are described in VHDL language. The physical realization is carried out on a RC100 Celoxica FPGA development board. [DOI]
A. Barriga, N. M. Hussein This chapter reviews image enhancement techniques. In particular the chapter is focused in soft computing technique to improve the contrast of images. There is a wide variety of contrast control techniques. However, most are not suitable for hardware implementation. A technique to control the contrast in images based on the application of Lukasiewicz algebra operators and fuzzy logic is described. In particular, the technique is based on the bounded-sum and the bounded-product. The selection of the control parameters is performed by a fuzzy system. An interesting feature when applying these operators is that it allows low cost hardware realizations (in terms of resources) and high processing speed. [DOI]
P. Brox, I. Baturone, S. Sánchez-Solano The tuning of hierarchical fuzzy systems are not supported by the majority of CAD tools available at the market currently. The xfsl tool integrated into Xfuzzy 3 allows the tuning of complex fuzzy systems, for instance, hierarchical systems with modules in cascade. The authors propose the use of this tool for tuning a complex fuzzy system for video deinterlacing in this paper. The parameters obtained after tuning are proven by de-interlacing a wide battery of sequences. The use of tuning techniques improves the quality of de-interlacing and provides an algorithm simplification that facilitates its hardware implementation. [DOI]
I. Baturone, A. Gersnoviez This paper describes a methodology to extract fuzzy models that describe linguistically the low-level features of an image (such as color, texture, etc.). The methodology combines grid-based algorithms with clustering and tabular simplification methods to compress image information into a small number of fuzzy rules with high linguistic meaning. All the steps of the methodology are carried out with the help offered by the tools of Xfuzzy 3 environment, so we can define, simplify, tune and verify the fuzzy models automatically. Several examples are included to illustrate the advantages of the methodology. [DOI]
N. M. Hussein, A. Barriga This paper describes a technique to control the contrast in images based on the application of Lukasiewicz algebra operators. In particular, the technique is based on the bounded-sum and the bounded-product. An interesting feature when applying these operators is that it allows low cost hardware realizations (in terms of resources) and high processing speed. The selection of the control parameters is perform by a fuzzy systems. [DOI]
N. M. Hussein, A. Barriga In this chapter a technique for detecting edges in images is presented. The technique is based on applying soft computing techniques such as fuzzy logic and Lukasiewicz algebra operator. The utility of this technique is related to the simplicity of the operations for edge calculation that makes it very suitable for low cost hardware implementation and high processing speed. [DOI]
P. Brox, I. Baturone, S. Sánchez-Solano A novel fuzzy motion-adaptive deinterlacing algorithm is presented in this paper. It uses fuzzy logic to interpolate between two processing modes, i.e., a spatial (IS) and a temporal (IT) interpolator. Furthermore, the temporal interpolator employs a very simple fuzzy inference system to implement a smart temporal interpolation that locally adapts to the features of the television (TV) material, such as possible picture repetition modes in the fields or in part of the fields (hybrid material). The combination of both systems provides effective results with a low cost in terms of computational resources. [DOI]
P. Brox, I. Baturone, S. Sánchez-Solano This paper presents an algorithm for video de-interlacing. The approach uses three fuzzy logic-based systems to adapt the interpolation strategy to the presence of motion and edges. Futhermore, the algorithm is able to deal with any kind of video material independently of the source used to acquire the scene. Extensive simulations of standard and real sequences prove the efficiency of the proposed algorithm.
P. Brox, I. Baturone, S. Sánchez-Solano
This paper surveys the hardware implementation of a de-interlacing algorithm on Field-Programmable
Technology for real-time processing. The algorithm presented evaluates the level of motion at each
pixel, and determines the interpolation between a spatial and a temporal method according to the
presence of motion. To achieve it the algorithm employs an hierarchical structure with three simple
fuzzy systems. The first one performs a set of fuzzy rules to apply reasoning in order to detect
motion; the second one selects the most convenient direction to implement an edge-dependent line
average method; and the third one is used to choose the most adequate temporal method.
P. Brox, I. Baturone, S. Sánchez-Solano A new motion adaptive algorithm for de-interlacing video is proposed in this paper. It employs two fuzzy systems to interpolate the missing lines of the transmission. One fuzzy system is used to evaluate the motion level at the current pixel, and a second one selects the most adequate temporal interpolation method. The combination of both systems provides an effective result with a low cost in term of hardware resources. [DOI]
P. Brox, I. Baturone, S. Sánchez-Solano Muchas tareas básicas de procesado de imágenes requieren la manipulación de grandes volúmenes de información que, en ocasiones, puede resultar ambigua y/o imprecisa como consecuencia de las características propias de las imágenes (gran cantidad de detalles con grandes contrastes de valores de luminancia y secuencias con un elevado grado de movimiento) o de los defectos de las mismas (presencia de ruido, falta de nitidez, etc.). En esta comunicación se analizan nuevas técnicas de interpolación basadas en lógica difusa que proporcionan soluciones eficaces para dos aplicaciones típicas de procesado de imágenes: el desentrelazado de señales de vídeo y el incremento de resolución de imágenes [Pdf]
P. Brox, I. Baturone, S. Sánchez-Solano, J. Gutiérrez-Ríos, F. Fernández-Hernández De-interlacing algorithms are required to convert interlaced video into progressive scan format. They perform an interpolation technique which doubles the vertical sampling density. This paper presents a de-interlacing algorithm which employs fuzzy logic to adapt the interpolation strategy to the presence of motion and edges. Extensive simulations of video sequences prove the advantages of this novel approach. [DOI]
P. Brox, I. Baturone, S. Sánchez-Solano
P. Brox, I. Baturone, S. Sánchez-Solano, J. Gutiérrez-Ríos, F. Fernández-Hernández A motion adaptive algorithm for video de-interlacing is presented in this paper. It is based on a fuzzy inference system, which performs an interpolation between two linear techniques as a function of the motion level. Fuzzy systems with different number of 'if-then' rules have been analyzed and compared in terms of complexity as well as efficiency in de-interlacing benchmark video sequences.
P. Brox, I. Baturone, S. Sánchez-Solano, J. Gutiérrez-Ríos, F. Fernández-Hernández En esta comunicación se presenta un algoritmo adaptativo con el movimiento para el desentrelazado de vídeo. Se basa en un sistema de inferencia difuso, que realiza una interpolación entre dos técnicas lineales en función del grado de movimiento. Se ha realizado un estudio de diferentes sistemas difusos con distinto número de funciones de pertenencia, analizándose el grado de complejidad de los mismos frente a su eficacia desentrelazando varias secuencias de vídeo. [Pdf]
P. Brox, I. Baturone, S. Sánchez-Solano The increase of resolution is one of the most important tasks in image processing. Traditional interpolation algorithms perform a linear interpolation between the closest pixels in the image. This strategy may introduce mistakes specially in the reconstruction of edges and zones with high contrast luminance values. The use of a novel interpolation algorithm for image enlargement is presented in this paper. It employs a fuzzy logic-system to adapt the interpolation to the presence of edges in the image, achieving good results at expense of a low increment in the computational cost. [Pdf]
P. Brox, I. Baturone, S. Sánchez-Solano Interlaced-to-progressive algorithms are currently required by video format conversion systems in order to display a progressive scanning used in modern visualization equipments. Deinterlacing algorithms use interpolation techniques to calculate missing pixels in transmitted fields. A motion adaptive algorithm which employs fuzzy logic to adapt the interpolation strategy to the presence of motion in the images is proposed in this paper. The performance of this new approach is evaluated by extensive simulation of different video sequences. [Pdf]
J. Gutiérrez-Ríos, F. Fernández, P. Brox, I. Baturone, S. Sánchez-Solano The procedure employed to make de-interlacing of video sequences has great influence in the quality of the obtained image. Reaching good results is not possible if dynamical characteristics of the processed image are not considered. On the other hand, the gradual adjust of the de-interlacing procedure as a function of the motion detected in each pixel of the image is a powerful method that is able to be realised by means of fuzzy inference. Detection of motion direction in each pixel of a frame becomes important in order to choose inclination in the spatial interpolation operations. In this paper we start from a fuzzy algorithm proposed by Van de Ville et al. to succeed in a family of more efficient algorithms under the point of view of execution speed and quality. These algorithms are based on convolution techniques (in substitution of the sum-prod norms) that are able to create a good emphasising distribution on the input variables. [Pdf]
P. Brox, S. Sánchez-Solano, I. Baturone, A. Barriga, J. Gutiérrez-Ríos, F. Fernández Los algoritmos de desentrelazado permiten la conversión de una señal de vídeo entrelazada a un formato progresivo. Entre las diferentes técnicas propuestas en la literatura, las que combinan interpolación espacial y temporal en función de la presencia de movimiento en la secuencia de imágenes proporcionan buenos resultados con un coste computacional bajo. Este artículo describe la implementación sobre FPGA de un algoritmo de desentrelazado que emplea lógica fuzzy para detectar el movimiento. Dependiendo del grado de movimiento evaluado en cada píxel se realiza la interpolación entre la técnica de inserción de campo y un nuevo método de promedio de línea dependiente de la detección de bordes que también utiliza lógica fuzzy. La realización de numerosas simulaciones con diferentes secuencias de vídeo muestran las ventajas del método propuesto frente a otras técnicas de desentralazado conocidas. La implementación del algoritmo sobre una FPGA proporciona una solución de bajo coste para procesado en tiempo real. [Pdf]
P. Brox, S. Sánchez-Solano, I. Baturone De-interlacing algorithms realize the interlaced to progressive conversion required in many applications. The most cost efficient are intra-field techniques, which interpolate pixels of the same field. Some of these methods use the upper and lower pixels lines. Among them, the ELA algorithm is widely employed since it reconstructs the edges of the de-interlaced image with more accuracy eliminating nondesired problems such as blurring and staircase effects. However, the ELA algorithm does not perform well when there are non clear edges or in presence of noise. In order to reduce these drawbacks, a new algorithm is presented in this paper. It is based on a simple fuzzy system which models heuristic rules to improve the ELA algorithm. Two enhancements of this new algorithm are also presented in this paper. Simulation results of video sequences prove the advantageous of the new algorithms. [Pdf]
P. Brox, S. Sánchez-Solano, I. Baturone De-interlacing methods realize the interlaced to progressive conversion required in many applications. Among them, intra-field methods are widely used for their good trade off between performance and computational cost. In particular, the ELA algorithm is well-known for its advantages in reconstructing the edges of the images, although it degrades the image quality where the edges are not clear. The algorithm proposed in this paper uses a simple fuzzy system which models heuristic rules to improve the ELA rules. It can be implemented easily in software and hardware since the increase in computational cost is very low. Simulation results are included to illustrate the advantages of the proposed fuzzy ELA algorithm in de-interlacing non noisy and noisy images. [Pdf]
P. Brox, S. Sánchez-Solano, I. Baturone, A. Barriga De-interlacing algorithms are used to convert interlaced video into progressive scan format. Among the different techniques reported in the literature, motion adaptive de-interlacing techniques that combine spatial and temporal interpolation according to the presence of motion achieve good results with a low computational cost. This paper presents the FPGA implementation of a motion adaptive algorithm which employs fuzzy logic in detecting motion and edges. Motion, which is evaluated at each pixel of the deinterlaced frame, determines the interpolation between an enhanced edge-dependent line average method and field insertion. Extensive simulations with video sequences show the advantages performance of the proposed method over other well-known de-interlacing techniques. The hardware implementation of the algorithm has been carried out on a FPGA obtaining a low-cost solution for real-time processing. [Pdf]
P. Brox, S. Sánchez-Solano, I. Baturone, A. Barriga Different architectures for hardware implementation of a de-interlacing video signal algorithm that combines spatial and temporal interpolation techniques are presented in this paper. The physical realization is carried out on an RC100 Celoxica board implementing a design described in VHDL language and generated with the tool System Generator (SysGen) from Xilinx. [Pdf]
| ||
|
Fuzzy Systems for Traffic Control in Internet Sistemas Difusos para Control de Tráfico en Internet |
||
|
F. Montesino, A. Barriga, D. R. López, S Sánchez-Solano Soft computing techniques and particularly fuzzy inference systems are gaining momentum as tools for network traffic modeling, analysis and control. Efficient hardware implementations of these techniques that can achieve real-time operation in high-speed networking equipment as well as other highly time-constrained application fields is however an open problem. We introduce a development platform for fuzzy inference systems with applications to network traffic analysis and control. The platform addresses the current requirements and constraints of high performance networking equipment. For the development process, we set up a methodology and a CAD tool chain that span the entire design process from initial specification in a high-level language to implementation on FPGA devices. An FPGA development board with PCI/PCIe interface is employed to support an open platform that comprises CAD tools as well as IP cores. PCI compatible fuzzy inference modules are implemented as System-on-Programmable-Chip (SoPC). We present satisfactory experimental results from the implementation of fuzzy systems for a number of applications in analysis and control of Internet traffic. These systems are shown to satisfy operational and architectural requirements of current and future high performance routing equipment. The platform proposed allows for the development of prototypes while avoiding large investments and complicated management procedures which constrain the testing and adoption of soft computing techniques in high performance networking. [DOI]
F. Montesino, A. Lendasse, A. Barriga We propose an automatic methodology framework for short- and long-term prediction of time series by means of fuzzy inference systems. In this methodology, fuzzy techniques and statistical techniques for nonparametric residual variance estimation are combined in order to build autoregressive predictive models implemented as fuzzy inference systems. Nonparametric residual variance estimation plays a key role in driving the identification and learning procedures. Concrete criteria and procedures within the proposed methodology framework are applied to a number of time series prediction problems. The learn from examples method introduced by Wang and Mendel (W&M) is used for identification. The Levenberg–Marquardt (L–M) optimization method is then applied for tuning. The W&M method produces compact and potentially accurate inference systems when applied after a proper variable selection stage. The L–M method yields the best compromise between accuracy and interpretability of results, among a set of alternatives. Delta test based residual variance estimations are used in order to select the best subset of inputs to the fuzzy inference systems as well as the number of linguistic labels for the inputs. Experiments on a diverse set of time series prediction benchmarks are compared against least-squares support vector machines (LS-SVM), optimally pruned extreme learning machine (OP-ELM), and k-NN based autoregressors. The advantages of the proposed methodology are shown in terms of linguistic interpretability, generalization capability and computational cost. Furthermore, fuzzy models are shown to be consistently more accurate for prediction in the case of time series coming from real-world applications. [DOI]
F. Montesino, A. Barriga, D. R. López, S. Sánchez-Solano We address, by means of fuzzy linguistic summaries, two related problems: summarizing network flow statistics and making these statistics human-readable. Two complementary summarization methods are developed. First, a fixed set of protoforms of interest is defined, and the ones with a higher truth value are shown to the user as simple on-line summaries. This first method is suitable for real-time monitoring. Then, an association rules mining process is carried out in order to find hidden relations in flow records. Both approaches are implemented in a tool capable of real-time and off-line processing of network flow records. Experimental results for a number of heterogeneous NetFlow records show the usefulness of linguistic summaries to both network practitioners and users. [DOI]
F. Montesino, A. Lendasse, A. Barriga We apply fuzzy techniques for system identification and supervised learning in order to develop fuzzy inference based autoregressors for time series prediction. An automatic methodology framework that combines fuzzy techniques and statistical techniques for nonparametric residual variance estimation is proposed. Identification is performed through the learn from examples method introduced by Wang and Mendel, while the Marquard-Levenberg supervised learning algorithm is then applied for tuning. Delta test residual noise estimation is used in order to select the best subset of inputs as well as the number of linguistic labels for the inputs. Experimental results for three time series prediction benchmarks are compared against LS-SVM based autoregressors and show the advantages of the proposed methodology in terms of approximation accuracy, generalization capability and linguistic interpretability. [DOI]
F. Montesino, A. Barriga, D.R. López, S. Sánchez-Solano Soft computing techniques are gaining momentum as tools for network traffic modeling, analysis and control. Efficient hardware implementations of these techniques that can achieve real-time operation in high-speed communications equipment is however an open problem. This paper describes a platform for the development of fuzzy systems with applications to communications systems, namely network traffic analysis and control. An FPGA development board with PCI interface is employed to support an open platform that comprises open CAD tools as well as IP cores. For the development process, we set up a methodology and a CAD tools chain that cover from initial specification in a high-level language to implementation on FPGA devices. PCI compatible fuzzy inference modules are implemented as SoPC based on the open WISHBONE interconnection architecture. We outline results from the design and implementation of fuzzy analyzers and regulators for network traffic. These systems are shown to satisfy operational and architectural requirements of current and future high-performance routing equipment.
F. Montesino, D. R. López, A. Barriga, S. Sánchez-Solano La dinámica del tráfico de extremo a extremo en Internet es un problema complejo para el cual los modelos disponibles son, en el mejor de los casos incompletos. Esta comunicación describe nuevos mecanismos para regulación de la tasa de transferencia de extremo a extremo en la capa de transporte por medio de sistemas difusos. Se describen una generalización basada en lógica difusa de los mecanismos de control de flujo y congestión de TCP (Transport Control Protocol), el diseño de un regulador difuso basado en mecanismo de ventana para TCP, así como la metodología de diseño empleada para simular e implementar de manera experimental el sistema. Se resume un estudio comparativo del regulador difuso presentado frente a los mecanismos tradicionales. El regulador difuso resulta útil como enfoque de modelado y proporciona significativas mejoras de prestaciones respecto a un conjunto de criterios. [Pdf]
F. Montesino, D. R. López, A. Barriga, S. Sánchez-Solano A major research problem in Internet transport protocols is the development of rate control mechanisms that can cope with a growing diversity of applications and services. We present novel mechanisms for intelligent end-to-end rate control in Internet by means of fuzzy systems. We outline a fuzzy generalization of TCP (Transport Control Protocol) rate control principles, and describe the design of a fuzzy TCP-like rate controller is then described. An evaluation of simulation an implementation results from the fuzzy rate controller as compared to that of traditional controllers is outlined. Besides being a useful modelling approach, the fuzzy rate controller is shown to outperform other approaches with regards to a number of criteria. [Pdf]
F. Montesino, D. R. López, A. Barriga, S. Sánchez-Solano A major research problem in Internet transport and network layers is the development of traffic regulation mechanisms that can cope with the requirements of a growing diversity of technologies, applications and services. This paper presents novel mechanisms for intelligent traffic scheduling in Internet routers by means of fuzzy logic based systems. A systematic design methodology, interpretability principles, evaluation over a broad range of network scenarios as well as practical implementation constraints have been considered. A comparative evaluation of results obtained by means of our fuzzy controllers as compared to that of traditional approaches is outlined. [Pdf]
F. Montesino, D. R. López, A. Barriga, S. Sánchez-Solano Los nuevos servicios, aplicaciones y tecnologías surgidos en Internet plantean serios problemas de gestión y optimización del uso del ancho de banda y otros recursos de la red. Se propone el uso de sistemas difusos como nuevo enfoque de análisis de los principales problemas y retos actuales en el diseño de mecanismos de control de congestión y calidad de servicio en Internet. En esta presentación se resumen tanto casos específicos de aplicación de este nuevo enfoque como futuros campos de aplicación, considerando el modelado de la red, los sistemas de medida y evaluación de prestaciones, y los sistemas de control de congestión y calidad de servicio. [Pdf]
|
||
|
Digital Implementation of Fuzzy Controllers Implementación Digital de Controladores Difusos |
||
|
S. Sánchez-Solano, A. J. Cabrera, M. Brox, A. J. González La continua demanda por parte del mercado microelectrónico de aplicaciones novedosas, con elevados niveles de complejidad y tiempos de desarrollo cortos ha motivado el impulso de las técnicas de diseño basadas en el concepto de "reusabilidad" y el desarrollo de elementos de sistemas como módulos de propiedad intelectual o módulos IP. En esta comunicación se describe la implementación de controladores difusos como módulos IP para FPGAs. Los controladores operan como periféricos conectables al bus OPB para los procesadores disponibles en las FPGAs de Xilinx. El empleo de las memorias internas de las FPGAs para almacenar las bases de conocimiento permite definir o ajustar la funcionalidad en tiempo de operación. [Pdf]
A. Barriga, S. Sánchez-Solano, P. Brox, A. Cabrera, I. Baturone The number of electronic applications using fuzzy logic-based solutions has increased considerably in the last few years. Concurrently, new CAD tools that explore different implementation technologies for this type of systems have been developed. In this paper we illustrate a fuzzy logic system design strategy based on a high level description. Employing this high level description, the knowledge base is translated to a format in appearance close to the natural language with the particularity that it uses a hardware description language (VHDL) directly synthesizable on an FPGA circuit. In addition, we analyze different approaches for FPGA implementations of fuzzy systems in order to characterize them in terms of area and speed. Among them, the use of specific processing architectures implemented on FPGAs presents as main advantages a good "cost-performance" ratio and an acceptably short development time. The different synthesis facilities provided by the Xfuzzy design environment for the implementation of programmable fuzzy systems, which take advantage of the available resources in the current FPGA families, are also analyzed in this paper. [DOI]
S. Sánchez-Solano, A. Barriga, P. Brox, I. Baturone En esta comunicación se presenta una estrategia de diseño para sistemas difusos basada en la utilización del lenguaje de descripción de hardware VHDL. Dicha estrategia consiste en aprovechar la flexibilidad que proporciona este lenguaje para describir los sistemas difusos utilizando un formato muy similar al lenguaje natural, y donde las definiciones de la estructura del sistema y de los operadores difusos se realizan por separado. La técnica utilizada permite diferenciar claramente el comportamiento funcional del sistema de los detalles de implementación, facilita las tareas de modelado de los sistemas y permite enlazar directamente con herramientas de síntesis automática de circuitos microelectrónicos. En la comunicación se analizan los resultados obtenidos mediante esta estrategia y se comparan con otros basados en arquitecturas específicas de los sistemas difusos. [Pdf]
A. Barriga, S. Sánchez-Solano, P. Brox, A. Cabrera, M. A. Marbán In this paper we will illustrate a fuzzy logic system design strategy based on high level description. Employing a high level description the knowledge base is described in a language appearance close to the natural language with the particularity of using a hardware description language (VHDL) directly synthesizable on an FPGA circuit. On the other hand we analyze FPGA implementations of different fuzzy inference hardware architectures in order to characteón permite expresar la base de conocimiento en un lenguaje muy próximo al lenguaje natural con la particularidad de realizarlo en un lenguaje de descripción de hardware (VHDL) directamente sintetizable en un dispositivo FPGA. Por otro lado también se realiza un análisis de implemenrize in terms of area and speed. [Pdf]
A. Barriga, M. A. Marbán, S. Sánchez-Solano, P. Brox, A. Cabrera En esta comunicación presentamos una estrategia de realización de sistemas difusos sobre FPGAs basada en una descripción de alto nivel. Dicha descripción permite expresar la base de conocimiento en un lenguaje muy próximo al lenguaje natural con la particularidad de realizarlo en un lenguaje de descripción de hardware (VHDL) directamente sintetizable en un dispositivo FPGA. Por otro lado también se realiza un análisis de implementaciones sobre FPGAs de diferentes arquitecturas hardware de sistemas de inferencia difusos con objeto de caracterizarlas (desde el punto de vista de área y velocidad). [Pdf]
S. Sánchez-Solano, A. Cabrera, C. J. Jiménez, P. Brox, I. Baturone, A. Barriga El número de aplicaciones electrónicas que utilizan soluciones basadas en lógica difusa se ha incrementado considerablemente en los últimos años y, de forma paralela, se han desarrollado nuevas herramientas de CAD que contemplan diferentes técnicas de implementación para este tipo de sistemas. De entre ellas, el uso de arquitecturas específicas de procesado implementadas sobre FPGAs presenta como principales ventajas una buena relación "coste-rendimiento" y un ciclo de desarrollo aceptablemente corto. En esta comunicación se analizan las distintas facilidades de síntesis que proporciona el entorno de diseño Xfuzzy para la implementación de sistemas difusos programables que aprovechen los recursos disponibles en las actuales familias de FPGAs. [Pdf]
A. Cabrera, S. Sánchez-Solano, C. J. Jimémez, A. Barriga, I. Baturone Se describen los elementos integrantes de una arquitectura de bajo costo y alto rendimiento para la implementación hardware de sistemas de inferencia difusos, la cual se basa en el procesado de reglas activas, la limitación del grado de solapamiento de las funciones de pertenencia de las entradas y la utilización de métodos de defuzificación simplificados. También se expone el entorno de desarrollo de sistemas difusos Xfuzzy, con énfasis en la herramienta xfvhdl, la cual permite la generación de código VHDL para los diferentes elementos de la arquitectura descrita. [Pdf]
A. Barriga, R. Senhadji, C. J. Jiménez, I. Baturone, S. Sánchez-Solano The main objective of this contribution is to present a design methodology for application specific fuzzy integrated circuits. This methodology is based on an specific architecture and a user-friendly design environment which enables the specification, verification and synthesis of fuzzy systems taking into account conceptual as well as microelectronics considerations.
E. Lago, M. A. Hinojosa, C. J. Jiménez, A. Barriga, S. Sánchez-Solano This paper explores the use of FPGA technologies to implement fuzzy logic controllers (FLCs). Two different approaches are described. The first option is based on the logic synthesis of the boolean equations describing the controller input-output relations. The second approach uses dedicated hardware to implement the fuzzy algorithm according to a specific architecture based on a VHDL cell library. In both alternatives, the synthesis process is accelerated by means of CAD tools which translate a high level description of the controller. A set of design examples are included in order to analyze the application domains covered by the different solutions. [Pdf]
S. Sánchez-Solano, A. Barriga, C. J. Jiménez, J. L. Huertas This paper focuses on hardware implementations of fuzzy inference systems which provide low silicon cost, high operational speed and adaptability to different application domains. The architecture and basic building blocks of two fuzzy logic controllers are described and their functionality is illustrated with experimental results showing the capability of these systems to be applied as function approximators.
C. J. Jiménez, S. Sánchez-Solano, A. Barriga La realización física de controladores difusos requiere la utilización de una arquitectura predeterminada, así como la selección de un conjunto de parámetros que configuren dicha arquitectura para la aplicación de control planteada. En esta comunicación se describe el desarrollo de una librería de celdas parametrizables para la síntesis de controladores difusos que aprovecha la potencialidad del VHDL como lenguaje de descripción de sistemas digitales. Las descripciones de las diferentes celdas cumplen, además, las restricciones impuestas por las principales herramientas de síntesis, lo que permite automatizar el diseño de este tipo de sistemas.
C. J. Jiménez, S. Sánchez-Solano, A. Barriga A general purpose, fuzzy logic-based controller is proposed. Its main features are low cost, high speed, and adaptability to different applications. The two first characteristics are obtained resorting to an active-rule driven strategy to implement simplified inference mechanism. Adaptability is insured by the capability to choose among different defuzzification methods. [Pdf]
C. J. Jiménez, S. Sánchez-Solano, A. Barriga, J. L. Huertas Se propone una clasificación de los distintos sistemas de inferencia basados en técnicas de razonamiento aproximado. Dicha clasificación distingue tres tipos de sistemas con diferentes dominios de aplicación y cuya implementación electrónica requiere el uso de distintas arquitecturas de procesado. Para el caso particular de los sistemas que denominaremos SISC se discuten una serie de alternativas de realización que permiten disminuir el área y aumentar su velocidad de inferencia, haciéndolos especialmente adecuados para aplicaciones de control en tiempo real. Finalmente, algunas de estas ideas se plasman en la realización en silicio de un prototipo que demuestra la viabilidad de las soluciones propuestas.
S. Sánchez-Solano, C. J. Jíménez, A. Barriga, J. L. Huertas Se propone una taxonomía que clasifica los sistemas de inferencia basados en lógica difusa de acuerdo con el carácter de la información que manejan en las diferentes etapas del proceso. Atendiendo a dicha taxonomía se revisan algunas de las implementaciones en hardware digital de sistemas difusos reportados en la literatura en la última década. Finalmente se proponen una serie de alternativas que permiten la realización eficiente de sistemas para aplicaciones de control en tiempo real.
C. J. Jiménez, A. Barriga, S. Sánchez-Solano A classification of inference systems based on approximate reasoning techniques is proposed. An alternative realization method is described for the particular SISC case, which enables reducing the silicon area and increasing the operation speed, making it especially appropriate for real time control applications. [Pdf]
| ||
| Mixed-Signal Implementation of Fuzzy Controllers Implementación de Controladores Difusos en Modo Mixto |
||
|
I. Baturone, S. Sánchez Solano Fuzzy controllers have been proven to be universal, that is, they can provide any control surface. Their microelectronic implementation is very suitable to achieve high-speed (real-time operation), and low area and power consumption. This paper focuses on discussing the two basic approaches that can be employed to design programmable universal controller integrated circuits. Analog, mixed-signal and digital realizations are summarized and compared. [Pdf]
I. Baturone, S. Sánchez-Solano, A. Barriga, C. J. Jiménez, R. Senhadji-Navarro, D. R. López Neuro-fuzzy systems can theoretically solve any problem since they are universal approximators. Besides, they combine the advantages of the neuro and fuzzy paradigms. This paper describes and compares the different strategies that can be adopted to implement the learning and inference mechanisms involved in a neuro-fuzzy system. CAD tools, most of them integrated into the fuzzy system development environment Xfuzzy 2.0 [1], have been developed to assist the designer in the implementation of neuro-fuzzy systems in FPGAs or ASICs. [Pdf]
I. Baturone, S. Sánchez Solano Fuzzy controllers have been proven to be universal, that is, they can provide any control surface. Their microelectronic implementation is very suitable to achieve high-speed (real-time operation), and low area and power consumption. This paper focuses on discussing the two basic approaches that can be employed to design programmable universal controller integrated circuits. Analog, mixed-signal and digital realizations are summarized and compared. [Pdf]
I. Baturone, G. Hernández, S. Sánchez-Solano VLSI implementation of universal approximator systems is very suitable to achieve high-speed (real-time operation), and low area and power consumption. This paper discusses three basic strategies to design programmable universal approximator ICs. Certain types of neural and fuzzy systems are included in these approximators. The circuitry required (digital, analog, and mixed-signal) and the approximation degree they achieve are discussed and compared. [Pdf]
I. Baturone, S. Sánchez Solano, J. L. Huertas Multiplier and divider circuits are usually required in the fields of analog signal processing and parallel-computing neural or fuzzy systems. In particular, this paper focuses on the hardware implementation of fuzzy controllers, where the divider circuit is usually the bottleneck. Multiplier/divider circuits can be implemented with a combination of A/D-D/A converters. An efficient design based on current-mode data converters is presented herein. Continuous-time algorithmic converters are chosen to reduce the control circuitry and to obtain a modular design based on a cascade of bit cells. Several circuit structures to implement these cells are presented and discussed. The one that is selected enables a better trade-off speed/power than others previously reported in the literature while maintaining a low area occupation. The resulting multiplier/divider circuit offers a low voltage operation, provides the division result in both analog and digital formats, and it is suitable for applications of low or middle resolution (up to 9 bits) like applications to fuzzy controllers. The analysis is illustrated with Hspice simulations and experimental results from a CMOS multiplier/divider prototype with 5-bit resolution. Experimental results from a CMOS current-mode fuzzy controller chip that contains the proposed design are also included. [Pdf]
I. Baturone, S. Sánchez Solano, A. Barriga, J. L. Huertas Several VLSI realizations of fuzzy systems have been proposed in the literature in the recent years. They employ analog or digital circuitry, offering more or less programmability, implementing different inference methods, with different types of membership functions as well as different antecedents' connectives. This paper centers this wide design space by fixing several parameters that allow efficient VLSI implementations of programmable fuzzy systems featuring first, second and third order accurate approximation. Hardware requirements are discussed and compared from the point of view of approximation capability or precision, thus attempting to a formalization that has never been applied before to the field of fuzzy hardware. [Pdf]
I. Baturone, S. Sánchez Solano, A. Barriga, J. L. Huertas A fuzzy processor is programmed to provide an optimum output for solving a given problem. It could theoretically solve any problem (from a static point of view) if it is an universal approximator. This paper addresses the design of fuzzy processors aiming at a twofold objective: efficient adaptive approximation of different and even dynamically changing surfaces and hardware simplicity. Adequate programmable parameters and a fully-parallel architecture are selected. Mixed-signal blocks based on digitally programmed current mirrors are employed. Error-descent learning algorithms for tuning are discussed. Adaptive behavior is illustrated with an application to the on-line identification of a nonlinear plant. [Pdf]
I. Baturone, S. Sánchez Solano, J. L. Huertas Combination of A/D-D/A converters allows implementing both long-term analog memory and multiplier/divider operations. An efficient design based on continuous-time, current-mode, dividing-algorithmic converters is presented in this paper. It offers high-speed and low-voltage operation with low or middle resolution. Experimental results of two CMOS-2.4µm prototypes (with 5- and 7-bit data converters) are provided. [Pdf]
I. Baturone, S. Sánchez-Solano, J. L. Huertas The required building blocks of CMOS fuzzy chips capable of performing as adaptive fuzzy systems are described in this paper. The building blocks are designed with mixed-signal current-mode cells that contain low-resolution A/D and D/A converters based on current mirrors. These cells provide the chip with an analog-digital programming interface. They also perform as computing elements of the fuzzy inference engine that calculate the output signal in either analog or digital formats, thus easing communication of the chip with digital processing environments and analog actuators. Experimental results of a 9-rule prototype integrated in a 2.4-µm CMOS process are included. It has a digital interface to program the antecedents and consequents and a mixed-signal output interface. The proposed design approach enables the CMOS realization of low-cost and high-inference fuzzy systems able to cope with complex processes through adaptation. This is illustrated with simulated results of an application to the on-line identification of a nonlinear dynamical plant. [Pdf]
I. Baturone, S. Sánchez-Solano, J. L. Huertas Combination of A/D-D/A converters allows implementing multiplier/divider operations. An efficient design based on continuous-time, current-mode, dividing-algorithmic converters is presented in this paper. It offers high speed and capability of low voltage operation, and it is suitable for applications of low or middle resolution (below 9 bits). In addition, the division result is given in both analog and digital formats. Experimental results from a CMOS prototype with 5 bit resolution are included.
I. Baturone, S. Sánchez-Solano, J. L. Huertas Hardware realization of adaptive fuzzy systems is discussed in this paper. A highly-parallel architecture based on an active-rule driven scheme is considered so that only the parameters associated with the active rules take part in the inference and learning phases. Mixed-signal circuits are used to implement a gradient-descent algorithm for tuning consequents' values and a weight perturbation method to adjust suitably chosen antecedents' parameters. Several applications of adaptive fuzzy chips are discussed showing that low-cost realizations are feasible.
I. Baturone, A. Barriga, S. Sánchez-Solano, J. L. Huertas This paper presents a novel architecture to implement general-purpose fuzzy chips. It allows fully-parallel rule processing employing a reduced number of mixed-signal computing blocks and minimum-sized digital memories. The resulting fuzzy processor can interact directly with continuous sensors and actuators and subsequent digital processing system.
I. Baturone, S. Sánchez-Solano, J. L. Huertas This paper presents a novel architecture to implement general-purpose fuzzy chips. It allows fully-parallel processing employing a reduced number of mixed-signal computing blocks and minimum-sized digital memories. The resulting fuzzy processor can interact directly with continuous sensors and actuators and subsequent digital processing systems. [Pdf]
I. Baturone, S. Sánchez-Solano, J. L. Huertas A long-term analogue memory based on continuous-time, current-mode algorithmic A/D-D/A converters is presented. It achieves an efficient performance in terms of speed, resolution, power, and area, while incorporating self-checking capability. Its operation is described and illustrated with experimental results of a CMOS-2.4µm prototype with 7-bit resolution.
I. Baturone, S. Sánchez-Solano, A. Barriga, J. L. Huertas The architecture and required building blocks of CMOS fuzzy chips capable of performing as adaptive fuzzy controllers are described in this paper. The building blocks are designed with mixed-signal current-mode cells that contain low-resolution A/D and D/A converters based on current mirrors. These cells provide the chip with analog or digital programmability and long-term local storage. They also perform as computing elements of the fuzzy inference engine that calculate the output signal in either analog or digital format, thus easing communication of the chip with digital processing environments and analog actuators. Experimental results of a 9-rule controller prototype integrated in a 2.4-µm CMOS process are included. It has a digital interface to program the antecedents and consequents and a mixed-signal output interface.
I. Baturone, S. Sánchez-Solano, A.Barriga, J. L. Huertas This paper discusses architectural and circuit-level aspects related to hardware realizations of fuzzy controllers. A brief overview on fuzzy inference methods is given focusing on chip implementation. The singleton or zero-order Sugeno's method is chosen since it offers a good trade-off between hardware simplicity and control efficiency. The CMOS microcontroller described herein processes information in the current-domain, but input-output signals are represented as voltage to ease communications with conventional control circuitry. Programming functionalities are added by combining analog and digital techniques, giving rise to a versatile microcontroller, capable of solving different control problems. After identifying the basic component blocks, the circuits used for their implementation are discussed and compared with other alternatives. This study is illustrated with the experimental results of prototypes integrated in different CMOS technologies.
J. L. Huertas, S. Sánchez-Solano, I. Baturone, A. Barriga This paper presents mixed-signal current-mode CMOS circuits to implement programmable fuzzy controllers that perform the singleton or zero-order Sugeno's method. Design equations to characterize these circuits are provided to explain the precision and speed that they offer. This ana-lysis is illustrated with the experimental results of prototypes integrated in standard CMOS technologies. These tests show that an equivalent precision of 6 bits is achieved. The connection of these blocks according to a proposed architecture allows fuzzy chips with low silicon area whose inference speed is in the range of 2 Mega FLIPS (fuzzy logic inferences per second).
I. Baturone, S. Sánchez-Solano, A. Barriga, J. L. Huertas This paper describes a programmable fuzzy controller chip designed with mixed-signal IC techniques. Its input and output signals are analog to directly interact with the information from the real world. The programmability interface is digital and the output signal is also given in digital format to allow easy embedding into digital processing environments. Experimental results from a prototype integrated in a 2.4-µm CMOS process are included.
S. Sánchez-Solano, I. Baturone, A. Barriga , J. L. Huertas En esta comunicación se presenta la arquitectura y los bloques funcionales de un controlador difuso CMOS de propósito general. Sus entradas y salidas son analógicas, por lo que puede interactuar directamente con la circuitería de control convencional. El sistema dispone de una interfaz digital que facilita su programación o la aplicación de técnicas de adaptación off-chip. Además permite incorporar procedimientos de ajuste on-chip que se realicen con circuitería digital o analógica. La utilización de técnicas de diseño en modo de corriente permite una realización poco costosa en términos de consumo de área y potencia.
J. L. Huertas, S. Sánchez-Solano, I. Baturone, A. Barriga This paper describes a system architecture for implementing fuzzy systems. Programming functionalities are added by combining analog and digital techniques, giving rise to a versatile microcontroller. Novel circuits to implement the required building blocks are provided and illustrated with experimental results. [Pdf]
I. Baturone, S. Sánchez-Solano, A. Barriga, J. L. Huertas This paper briefly reviews different inference/defuzzification methods reported in the literature and concludes with the best approaches in terms of control and hardware efficiency. Three strategies based on analog circuits are presented for their implementation. They are discussed at an architectural and operator level taking into account deterministic and random errors, area, power and frequency behavior. [Pdf]
S. Sánchez-Solano, I. Baturone, A. Barriga, J. L. Huertas This paper presents a versatile current-mode circuit which combines the advantages of analog and digital techniques to perform long-term memory and computing functions. It is based on current-mode algorithm data converters and standard semistatic latches. Hspice simulations are included to illustrate its behavior. The design of the A/D subcell has been improved by using current injection techniques and enhanced current comparators. The cell can be employed as a building block to implement programmable fuzzy chips, introduce learning in so-called adaptive neuro-fuzzy chips, or develop fuzzy sequential processors as an extension of ordinary binary computers. [Pdf]
I. Baturone, S. Sánchez-Solano, A. Barriga, J. L. Huertas La división es una operación básica en la etapa final de los controladores difusos (fase de defuzzificación). En este trabajo se analizan estructuras simples de divisores basados en transistores MOS operando en inversión fuerte y se discuten sus características en cuanto a capacidad de interfaz con los circuitos precedentes y posteriores, rango dinámico, velocidad de operación, precisión del comportamiento y consumo de potencia y área. Como resultado de este estudio, proponemos la utilización de una de estas estructuras para el bloque defuzzificador, que ofrece más ventajas que las soluciones reportadas.
I. Baturone, S. Sánchez-Solano, A. Barriga, J. L. Huertas Se describe la realización de un controlador difuso que implementa el mecanismo de inferencia simplificado utilizando circuitos analógicos CMOS en los que las señales son representadas mediante intensidades. No obstante, el sistema completo se comunica con el exterior mediante tensiones, con objeto de permitir una interfaz sencilla con la circuitería de control convencional. Se discute el método de inferencia empleado, así como distintas alternativas para la realización de los circuitos de generación de los antecedentes, los operadores Min/Max y la etapa de ponderación de reglas. La elección de las soluciones más eficaces para cada bloque del controlador permite construir un sistema óptimo en términos de velocidad de operación y consumo de área y potencia.
J. L. Huertas, S. Sánchez-Solano, A. Barriga, I. Baturone A new architecture is presented for the implementation of fuzzy systems using analog-digital techniques. This architecture is directed towards allowing the implementation of many rules on the same chip, including the fuzzy inference engine and the defuzzifier. This approach is based on a total or partial sequential operation of both the fuzzifier and the defuzzifier. A basic operational cell for a membership function circuit (MFC) as well as its programmable version are described and used for realizing the proposed architecture in a CMOS technology. Using a similar structure to that of the MFCs basic cell, new implementations for MIN-MAX operators are also presented. [Pdf]
I. Baturone, S. Sánchez-Solano, J. L Huertas This paper describes the realization of a fuzzy controller which implements the simplified inference mechanism. CMOS analog circuits where signals are represented as currents are employed. However, the whole system is externally communicated through voltages, thus enabling simple interface with conventional control circuitry. [Pdf]
I. Baturone, A. Barriga, J. L Huertas This paper presents briefly a review of the different algorithms and hardware implementations for multi-input min/max operators. This allow us to show how a clever utilization of MOS devices leads to very simple current and voltage-mode min/max circuits. [Pdf]
I. Baturone, S. Sánchez-Solano, A. Barriga, J. L.Huertas This paper explores the solutions offered by Switched Capacitor (SC) techniques for the implementation of fuzzy circuits. The architecture of a microcontroller is presented to perform the simplified inference method and the circuit realizations of the different blocks are described. The design technique achieves high operation speed, typical in analog operation, and at the same time allows the programming ease which is characteristic of digital implementations. Finally, due to the defuzzifier used, the proposed microcontroller may interact with either an analog or digital system at the output.
I. Baturone, J. L. Huertas, A. Barriga, S. Sanchez-Solano A CMOS continuous time current-mode circuit which provides the maximum of n analog inputs is presented. This structure exhibits a O(n) complexity, and allows high precision and speed with small area and very low power dissipation. Its operation is discussed and illustrated with simulation results.
I. Baturone, S. Sánchez-Solano, A. Barriga, J.L. Huertas En esta comunicación se exploran las soluciones que ofrecen las técnicas de condensadores en conmutación (SC) para la realización hardware de sistemas difusos. Se presenta la arquitectura de un microcontrolador que implementa el método de inferencia simplificado y se describen las realizaciones de circuito de los diferentes bloques constitutivos. La técnica de diseño adoptada proporciona una elevada velocidad de operación, típica de las realizaciones analógicas, al tiempo que permite la facilidad de programación característica de las implementaciones digitales. Por último, como consecuencia del esquema utilizado, el microcontrolador propuesto puede interactuar, a la salida, con un sistema tanto analógico como digital.
S. Sánchez-Solano, I. Baturone, A. Barriga, J.L. Huertas La realización en hardware de sistemas de control basados en Lógica Difusa ha experimentado una continua evolución en los últimos años. La búsqueda de soluciones óptimas desde la perspectiva microelectrónica ha motivado la aparición de una serie de nuevas realizaciones que difieren entre sí tanto en aspectos arquitecturales como de implementación. En este artículo se discuten las diferentes alternativas por las que puede optar el diseñador de circuitos VLSI y se analiza la evolución de los controladores difusos mediante el estudio de una serie de propuestas reportadas por los autores en los últimos años.
J. L. Huertas, S. Sánchez-Solano, I. Baturone, A. Barriga A fuzzy microcontroller is presented implementing a simplified inference mechanism. Fuzzification, rule composition and defuzzification are carried out by means of (basically) analog current-mode CMOS circuits operating in strong inversion. Also a voltage interface is provided with the external world. Combining analog and digital techniques allow a programming capability. [Pdf]
J. L. Huertas, S. Sánchez-Solano, A. Barriga, I. Baturone The use of Switched-Capacitor techniques to build a fuzzy controller is discussed in this contribution. Using a sequential architecture, the required building blocks are introduced and its realization is described. The proposed system can be considered as a starting point for exploring the future capabilities offered by SC networks to the hardware implementations of fuzzy systems.
I. Baturone, A. Barriga, S. Sánchez-Solano, J. L. Huertas El creciente interés despertado por los circuitos que operan en lógica difusa nos conduce a revisar las diferentes técnicas utilizadas en la implementación de los bloques básicos de estos sistemas. Esta comunicación plantea un nuevo procedimiento para generar las funciones de pertenencia que representan a los conjuntos difusos. A partir de esta técnica, se describe la realización de una celda operacional básica, así como una versión de la misma que incluye programabilidad digital. Ambos circuitos han sido implementados en una tecnología CMOS.
J. L. Huertas, S. Sánchez-Solano, A. Barriga, I. Baturone A new architecture is presented for the implementation of fuzzy systems using analog-digital techniques. This architecture is directed towards allowing the implementation of many rules on the same chip, including the fuzzy inference engine and the defuzzifier. This approach is based on a total or partial sequential operation of both the fuzzifier and the defuzzifier. A basic operational cell for a membership function circuit as well as its programmable version are described and used for realizing the proposed architecture in a CMOS technology. [Pdf]
|
||
|
©IMSE-CNM 2012 |
