2001, Vol. VIII, Núm. 2
http://hdl.handle.net/2099/2073
Thu, 15 Aug 2024 11:50:17 GMT2024-08-15T11:50:17ZA multicriteria genetic tuning for fuzzy logic controllers
http://hdl.handle.net/2099/3604
A multicriteria genetic tuning for fuzzy logic controllers
Alcalá Fernández, Rafael; Casillas Barranquero, Jorge; Castro Peña, Juan Luis; González Muñoz, Antonio; Herrera Triguero, Francisco
This paper presents the use of genetic algorithms to develop smartly tuned fuzzy logic controllers in multicriteria complex problems. This tuning approach has some specific restrictions that make it very particular and complex because of the large time requirements existing due to the need of considering multiple criteria ---which enlarges the solution search space---, and to the long computation time models usually used for fitness assessment. To solve these restrictions, two efficient genetic tuning strategies considering different multicriteria approaches have been developed and tested in a real-world problem for fuzzy control of HVAC Systems.
Mon, 01 Oct 2007 09:51:23 GMThttp://hdl.handle.net/2099/36042007-10-01T09:51:23ZAlcalá Fernández, RafaelCasillas Barranquero, JorgeCastro Peña, Juan LuisGonzález Muñoz, AntonioHerrera Triguero, FranciscoThis paper presents the use of genetic algorithms to develop smartly tuned fuzzy logic controllers in multicriteria complex problems. This tuning approach has some specific restrictions that make it very particular and complex because of the large time requirements existing due to the need of considering multiple criteria ---which enlarges the solution search space---, and to the long computation time models usually used for fitness assessment. To solve these restrictions, two efficient genetic tuning strategies considering different multicriteria approaches have been developed and tested in a real-world problem for fuzzy control of HVAC Systems.A multi-agent system based on fuzzy logic applied to RoboCup's enviroment
http://hdl.handle.net/2099/3603
A multi-agent system based on fuzzy logic applied to RoboCup's enviroment
Aguirre Molina, Eugenio; Gámez, Juan Carlos; González Muñoz, Antonio
Artificial Intelligent has been applied successfully to many and
varied domains. In particular, the recent steps forward which are
been producing in the development of dynamic, collaborative, real
time and adversarial multi-agent environments are very
interesting. In this work, we carry out a proposal, which uses a
soccer robot simulator, called TeamBots, that simulates the
``RoboCup Small'', involved into the ``RoboCup World
Championship''. The objective is to bring a team into operation,
even though, this team initially only has available the basic
behaviours needed to be operative and able to participate in a
championship with other teams. So, we are proposed ourselves the
development of the complete team, and subsequently in the future
to go improving its capabilities and possibilities. In this work,
we show the team's design in broad outlines, and we explain in
detail the strategy of game for the different players, displaying
moreover a part of the experimentation that has been carried out
to state the correct operation of the team.
Mon, 01 Oct 2007 09:39:08 GMThttp://hdl.handle.net/2099/36032007-10-01T09:39:08ZAguirre Molina, EugenioGámez, Juan CarlosGonzález Muñoz, AntonioArtificial Intelligent has been applied successfully to many and
varied domains. In particular, the recent steps forward which are
been producing in the development of dynamic, collaborative, real
time and adversarial multi-agent environments are very
interesting. In this work, we carry out a proposal, which uses a
soccer robot simulator, called TeamBots, that simulates the
``RoboCup Small'', involved into the ``RoboCup World
Championship''. The objective is to bring a team into operation,
even though, this team initially only has available the basic
behaviours needed to be operative and able to participate in a
championship with other teams. So, we are proposed ourselves the
development of the complete team, and subsequently in the future
to go improving its capabilities and possibilities. In this work,
we show the team's design in broad outlines, and we explain in
detail the strategy of game for the different players, displaying
moreover a part of the experimentation that has been carried out
to state the correct operation of the team.A formal analisys of the computational dynamics in GIGANTEC
http://hdl.handle.net/2099/3602
A formal analisys of the computational dynamics in GIGANTEC
Badr, Amr
An evolutionary algorithm formalism has been forwarded in a previous research, and implemented in the system GIGANTEC: Genetic Induction for General Analytical Non-numeric Task Evolution Compiler [Bad98][Bad99]. A dynamical model is developed to analyze the behaviour of the algorithm. The model is dependent in its analysis on classical Compilers Theory, Game Theory and Markov Chains and its convergence characteristics. The results conclude that a limiting state is reached, which is independent of the initial population and the mutation rate, but dependent on the cardinality of the alphabet of the driving L-system.
Mon, 01 Oct 2007 09:22:09 GMThttp://hdl.handle.net/2099/36022007-10-01T09:22:09ZBadr, AmrAn evolutionary algorithm formalism has been forwarded in a previous research, and implemented in the system GIGANTEC: Genetic Induction for General Analytical Non-numeric Task Evolution Compiler [Bad98][Bad99]. A dynamical model is developed to analyze the behaviour of the algorithm. The model is dependent in its analysis on classical Compilers Theory, Game Theory and Markov Chains and its convergence characteristics. The results conclude that a limiting state is reached, which is independent of the initial population and the mutation rate, but dependent on the cardinality of the alphabet of the driving L-system.Fuzzy set theory for cumulative trauma prediction
http://hdl.handle.net/2099/3601
Fuzzy set theory for cumulative trauma prediction
Fonseca, Daniel J.; Merritt, Thomas W.; Moynihan, Gary P.
A widely used fuzzy reasoning algorithm was modified and implemented via an expert system to assess the potential risk of employee repetitive strain injury in the workplace. This fuzzy relational model, known as the Priority First Cover Algorithm (PFC), was adapted to describe the relationship between 12 cumulative trauma disorders (CTDs) of the upper extremity, and 29 identified risk factors. The algorithm, which finds a suboptimal subset from a group of variables based on the criterion of priority, was adopted to enable the inference mechanism of a constructed knowledge-based system to predict CTD occurrence.
Mon, 01 Oct 2007 09:12:14 GMThttp://hdl.handle.net/2099/36012007-10-01T09:12:14ZFonseca, Daniel J.Merritt, Thomas W.Moynihan, Gary P.A widely used fuzzy reasoning algorithm was modified and implemented via an expert system to assess the potential risk of employee repetitive strain injury in the workplace. This fuzzy relational model, known as the Priority First Cover Algorithm (PFC), was adapted to describe the relationship between 12 cumulative trauma disorders (CTDs) of the upper extremity, and 29 identified risk factors. The algorithm, which finds a suboptimal subset from a group of variables based on the criterion of priority, was adopted to enable the inference mechanism of a constructed knowledge-based system to predict CTD occurrence.Rule-based fuzzy object similarity
http://hdl.handle.net/2099/3600
Rule-based fuzzy object similarity
Bunke, Horst; Fàbregas, Xavier; Kandel, Abraham
A new similarity measure for objects that are represented by
feature vectors of fixed dimension is introduced. It can
simultaneously deal with numeric and symbolic features.
Also, it can tolerate missing feature values. The similarity measure between
two objects is described in terms of the similarity of their
features. IF-THEN rules are being used to model the
individual contribution of each feature to the global similarity measure
between a pair
of objects. The proposed similarity measure is based on
fuzzy sets and this allows us to deal with vague, uncertain and distorted
information in a natural way. Several formal properties of the
proposed similarity measure are derived; in particular, we show that the measure can be used to model the Euclidean
distance as well as other, non-Euclidean distance functions. Also, an
application of the proposed similarity measure to
nearest-neighbor classification in a medical expert system is
described.
Mon, 01 Oct 2007 08:58:49 GMThttp://hdl.handle.net/2099/36002007-10-01T08:58:49ZBunke, HorstFàbregas, XavierKandel, AbrahamA new similarity measure for objects that are represented by
feature vectors of fixed dimension is introduced. It can
simultaneously deal with numeric and symbolic features.
Also, it can tolerate missing feature values. The similarity measure between
two objects is described in terms of the similarity of their
features. IF-THEN rules are being used to model the
individual contribution of each feature to the global similarity measure
between a pair
of objects. The proposed similarity measure is based on
fuzzy sets and this allows us to deal with vague, uncertain and distorted
information in a natural way. Several formal properties of the
proposed similarity measure are derived; in particular, we show that the measure can be used to model the Euclidean
distance as well as other, non-Euclidean distance functions. Also, an
application of the proposed similarity measure to
nearest-neighbor classification in a medical expert system is
described.Marginalization like a projection
http://hdl.handle.net/2099/3599
Marginalization like a projection
Verdegay López, Juan Francisco; Moral Callejón, Serafín
This paper studies the problem of marginalizing convex polytopes of probabilities
represented by a set of constraints. This marginalization is obtained as a special
case of projection on a specific subspace. An algorithm that projects a convex
polytope on any subspace has been built and the expression of the subspace, where the
projection must be made for obtaining the marginalization, has been calculated.
Mon, 01 Oct 2007 08:40:35 GMThttp://hdl.handle.net/2099/35992007-10-01T08:40:35ZVerdegay López, Juan FranciscoMoral Callejón, SerafínThis paper studies the problem of marginalizing convex polytopes of probabilities
represented by a set of constraints. This marginalization is obtained as a special
case of projection on a specific subspace. An algorithm that projects a convex
polytope on any subspace has been built and the expression of the subspace, where the
projection must be made for obtaining the marginalization, has been calculated.Averaging Premises
http://hdl.handle.net/2099/3598
Averaging Premises
Trillas i Gay, Enric; Castiñeira Holgado, Elena Esther; Cubillo Villanueva, Susana
This paper deals with the sets of strict conjectures and
consequences of a given collection $P$ of premises. The set of
Averaging Functions is introduced on lattices and some properties
of these functions are shown. Averaging Functions allow to
interpret "restricted consequences'' as averages of premises. The
subset of consequences $C^\ast_g(P)$ and the subset of conjectures
$\Phi^\ast_g(P)$ defined by means of the averaging function $g$
are introduced, and their properties are studied. This sets allow
to give decomposition theorems for the restricted consequences and
for the strict conjectures.
Fri, 28 Sep 2007 11:41:52 GMThttp://hdl.handle.net/2099/35982007-09-28T11:41:52ZTrillas i Gay, EnricCastiñeira Holgado, Elena EstherCubillo Villanueva, SusanaThis paper deals with the sets of strict conjectures and
consequences of a given collection $P$ of premises. The set of
Averaging Functions is introduced on lattices and some properties
of these functions are shown. Averaging Functions allow to
interpret "restricted consequences'' as averages of premises. The
subset of consequences $C^\ast_g(P)$ and the subset of conjectures
$\Phi^\ast_g(P)$ defined by means of the averaging function $g$
are introduced, and their properties are studied. This sets allow
to give decomposition theorems for the restricted consequences and
for the strict conjectures.(┬,┴,N) Fuzzy logic
http://hdl.handle.net/2099/3597
(┬,┴,N) Fuzzy logic
Xu, Y.; Liu, J.; Ruan, Da
To investigate more reasonable fuzzy reasoning model in expert
systems as well as more effective logical circuit in fuzzy
control, a $(\top, \bot, N)$ fuzzy logic is proposed in this paper
by using $\top$-norm, $\bot$-norm and pseudo-complement $N$ as
the logical connectives. Two aspects are discussed: (1) some
concepts of ($\top, \bot, N)$ fuzzy logic are introduced and some
properties of ($\top, \bot, N)$ fuzzy logical formulae are
discussed. (2) $G$-fuzzy truth (falsity) of $(\top, \bot, N)$
fuzzy logical formulae are investigated and also the relation
between the Boolean truth (falsity) of $\bot$-normal forms
($\top$-normal forms) and the $G$-fuzzy truth (falsity) of them
are
analyzed.\
Fri, 28 Sep 2007 11:29:38 GMThttp://hdl.handle.net/2099/35972007-09-28T11:29:38ZXu, Y.Liu, J.Ruan, DaTo investigate more reasonable fuzzy reasoning model in expert
systems as well as more effective logical circuit in fuzzy
control, a $(\top, \bot, N)$ fuzzy logic is proposed in this paper
by using $\top$-norm, $\bot$-norm and pseudo-complement $N$ as
the logical connectives. Two aspects are discussed: (1) some
concepts of ($\top, \bot, N)$ fuzzy logic are introduced and some
properties of ($\top, \bot, N)$ fuzzy logical formulae are
discussed. (2) $G$-fuzzy truth (falsity) of $(\top, \bot, N)$
fuzzy logical formulae are investigated and also the relation
between the Boolean truth (falsity) of $\bot$-normal forms
($\top$-normal forms) and the $G$-fuzzy truth (falsity) of them
are
analyzed.\