Top Banner

of 50

Research Report 2006

Apr 05, 2018

Download

Documents

annafeliu90
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • 7/31/2019 Research Report 2006

    1/50

    ResearchReport2006

    UdG

  • 7/31/2019 Research Report 2006

    2/50

    Research Report 2006

  • 7/31/2019 Research Report 2006

    3/50

    ContentContent

    1. History and Objectives

    2. Organisation and Management3. Research Groups4. Public-Funded Projects

    4.1. GameTools - Advanced Tools for Developing Highly Realistic Computer Games4.2. MOMARnet - Marie Curie Research Training Network - Monitoring the Mid-Atlantic Ridge4.3. FREESUBNET - European Research network on key technologies for intervention autonomous underwater vehicles4.4. ONE - Open Negotiation Envirnoment4.5. Fault Diagnosis of Electrical Distribution Networks Based on Cases and Models4.6. Automated fault detection, diagnosis and tolerance in uncertain systems and distributed systems4.7. Multilayer Mechanisms for Resource allocation and Restorable Routing in GMPLS-based core networks

    4.8. NetMould. Efficient technology integration of the mould in knowldege based networks4.9. Development of a prototype to assist fault location in distribution networks4.10. Reference network in Advanced Production Techniques4.11. Analysis, identification and control of mechatronic systems with hysteresis and friction. Applicationto piezoelectric and magnetorheological actuators4.12. AIRSUB - Underwater Robotics Industrial Application4.13. Virtual Prototyping for Architectural Projects4.14. Developing Distributed Computing Techniques for Scheduling. Application to Transport Problems4.15. Interfaces avanzadas para la interaccin transparente con modelos altamente complejos en entornos de realidad virtual4.16. AdaptaPlan - Adaptation based on machine learning, user modelling and planning for complex

    useroriented tasks4.17. Improvement and Application of the Game Technology in Virtual Reality and web contents4.18. Algorithms and Graphical Interfaces for Applications Based on Natural and Artificial LightingSimulation4.19. INSULAID - Development of a glucemic prediction and insulin dosage-aid system for patientswith diabetes mellitus4.20. Development of an Intelligent Control System applied to a Sequencing Batch Reactor(SBR) for the removal of Organic Matter, Nitrogen and Phosphorous4.21. Development of a Modular System to Build Georeferenced Photo-mosaics of the Ocean Floor4.22. Management of circuit-switched IP networks (R-MPLS)

    5. Technology Transfer5.1. Endesa5.2. Optimus

    6. Education7. Competition8. Publications

  • 7/31/2019 Research Report 2006

    4/50

    4 Research Report 2006

  • 7/31/2019 Research Report 2006

    5/50

    1. History and Objectives1. History and Objectives

    History

    The research Institute for Informatics and Applied Matematics (IIiA) from University of Girona was created in 1996.It currently has more than 120 members, including 45 doctors. The researchers run their activities in different

    fields such as computer graphics, computer vision, robotics, telecomunications, automatics, control engineering,

    and artificial intelligence. They are members of two departments from the University of Girona: the Department

    of Electronics, Computer Engineering and Automatics and the Department of Informatics and Applied Mathematics.

    Objectives

    The main purpose of the IIiA is to acquire an international reputation in its scientific and technical activity. This re-

    putation has its basis in the conception, design and implementation of R&D projects and technology transfer con-

    tracts. These activities produce innovative results in all their phases, both in scientific and engineering terms.

    Moreover, the Institute aims at contributing to the economic generation and growth of its region industrial contextin the Information Technologies sector.

    The IIiA specific objectives are:

    To promote research activities, technology transfer and specialized teaching in the fields of Infor-mation Technologies and its applications.

    To foster and coordinate interdisciplinary projects using the different areas of expertise of the insti-tute members.

    To encourage the development of pre-competitive projects and its materialization in technology in-novation contracts.

    To motivate the participation of the University of Girona in consortiums that focus in research linesclose to the ones of its members. The Institute acts as the representative of the University in suchconsortiums.

    5Research Report 2006

  • 7/31/2019 Research Report 2006

    6/50

    2. Organization and Management2. Organization and Management

    Organization

    The IIiA comprises 5 research groups:

    Broadband Communications and Distributed Systems (BCDS)

    Control Engineering and Intelligent Systems (eXiT)

    Girona Graphics Group (GGG)

    Modal Intervals and Control Engineering (MICE)

    Computer Vision and Robotics (Vicorob)

    Management

    The Council is the most important management organ, and it is composed by all the academic staff of the Institute.

    The Directoris the head of the Institute. It is elected after an open selection process.

    An Executive Commission represents the council and gives advice to the Director in government tasks. It is for-

    med by the Director, the head of each research group, the academic secretary and the research manager.

    The Research Manageris responsible for the internal operation of the Institute.

    The Academic Secretary is in charge of writing the minutes of the Council and Executive Commission meetings

    and write certificates of the agreements reached on them.

    Director: Dr. Josep Veh

    Head of BCDS: Dr. Josep Lluis Marzo

    Head of eXiT: Dr. Joaquim Melendez

    Head of GGG: Prof. Xavier Pueyo

    Head of MICE: Dr. Josep Veh

    Head of Vicorob: Dr. Joan Mart

    Research Manager: Mr. Jordi Palau

    Academic Secretary: Dr. Marc Carreras

    The Secretariat of the Institute provides administrative support to the Institute Members.

    Ms. Rosa Teixidor

    Ms. Maria Brugu

    6 Research Report 2005

  • 7/31/2019 Research Report 2006

    7/50

    3. Research Groups3. Research Groups

    Broadband Communications and Distributed Systems (BCDS)

    The BCDS group (ref. GRCT40) is active in two main research lines: broadband networks management and con-trol and distributed systems. The group has led several long-term governmental projects involving other universities

    and companies like Internet Service Providers and Public Health Institutions. These projects, along with the basic

    research performed on the group, bring about many publications and contributions on international journals,

    symposiums and conferences. BCDS os part of the DURSI consolidated research group AEDS (ref. SGR-00296).

    The following are BCDS current research topics:

    Broadband Network Control and Management

    Control and Management of Quality of Service (QoS) networks

    Multi-Agent Systems in Distributed Network Management

    Interdomain Management and Virtual Private Networks

    Optimization Techniques (Ant Colony Opt., Multi-objective Genetic Algorithms, Linear Programming,

    etc)

    Connection Admission Control Mechanisms for elastic (TCP) traffic.

    Dynamic Resource Allocation and Load Balancing.

    Charging for Multicast QoS Services

    Multi-level ([G]MPLS & optical) Routing and Protection mechanisms

    Distributed Applications

    Distributed Processing and Parallelism (Grids)

    Distributed Artificial Intelligence: Multi-Agent Systems

    Open and Distance Learning

    Adaptive Models for Collaborative Distance Learning

    Dynamic Adaptive Hypermedia Systems

    BCDS website: http://bcds.udg.es

    Control Engineering and Intelligent Systems

    This group is divided in two different research groups: eXiT and ARLab.

    eXiT is a research group (GRCT 41) of the Universitat de Girona integrated by professors and phD. students of

    the Departament d'Electrnica Informtica i Automtica . The research activity is developed in the Institut d'Infor-

    mtica i Aplicacions . The eXiT teaching capacity is developed in the IT Doctoral program (Dept. EIA / IGSOC).

    7Research Report 2006

  • 7/31/2019 Research Report 2006

    8/50

    eXiT group develops an active research in the field of process supervision with emphasis in the integration of me-thods and techniques to assess process behaviour from measurements and experience reuse based on the fo-llowing main points:

    Case Based Diagnosis: Adaptation of Case Based Reasoning (CBR) methodology into the domain

    of dynamic process with emphasis on diagnosis. Definition of similarity criteria between signals.Reuse of past experiences.

    Multivariate statistical process control: adaptation of these formalisms to model, fault detection anddiagnosis of batch process.

    Qualitative representation of trends: definition, formalization and development of numeric to quali-tative interfaces. Representation of signals as sequences of episodes necessary to deal with kno-wledge based approaches.

    Knowledge discovering in data bases for situation assessment.

    This research is clearly domain dependent. Nowadays, the activity is developed in the following domains:

    Chemical and petrochemical process Waste water treatment plants

    Electric Distribution System (Power Quality)

    Electronic Circuits

    Other recent research interest is in the field of Remote and Virtual laboratories. This is motivated by the necessityof testing and comparing experiences in shared configurable process and the teaching activity of the group.

    eXiT website: http://exit.udg.es

    Agents Research Lab is a research group (GRCT 69) that is concerned with the development and analysis of AI

    techniques and control architectures for both agents and multiagent systems. The origin of our group is the ap-plication of Artificial Intelligence to control and supervision. In these issues, the group has obtained successful re-sults in the development of physical agents. Such agents have specific properties provided that they have a bodythat interact with the environment and need to adapt their behaviour to the new circumstances they encounter. Cu-rrently, we are exploiting these particular properties of physical agents to personalised agents. This involves, re-search on multi-agent architectures for user modelling and recommender systems, trust mechanisms for agentcollaboration, integration information frameworks and social networks. The group collaborates in innovation pro-jects in the industry within the EASY Center (CIDEM/X-IT network).

    ARLab website: http://eia.udg.es/arl

    Girona Graphics Group (GGG)

    The GGG began its research activity at 1993 with the study of problems related to the realistic images generationsoftware. Since then, the GGG has extended the focus of its work dealing with several aspects of visualizationfrom both scientific and technological point of view. Thus, actually the GGG develops his research in:

    Realistic rendering

    Scientific visualization

    Computional geometry

    Appearance and perception

    Virtual Reality

    In these research lines, the GGG deals with problems related to interactive graphic and visualization fundamentals,

    8 Research Report 2006

  • 7/31/2019 Research Report 2006

    9/50

    as well as with applications of diverse types :

    Medical imaging

    Public lighting

    Lighting design Luminaries design

    Driving simulation

    Interactive games

    GGG website: http://iiia.udg.es/GGG

    Modal Intervals and Control Engineering (MICE)

    The imprecision and the uncertainty of complex systems can be represented in many cases by interval models.

    The core of our research is to develop methods and tools based on interval analysis to deal with this kind ofsystems.

    Research areas:

    Theoretic development of Modal Interval Analysis

    Management of interval uncertainties

    Interval Simulation

    Linear interval systems

    Development of tools for interval computations

    Modal Interval Libraries

    Minimax and optimization algorithms

    Quantified Set Inversion

    Application of interval Methods to Systems and Control

    Robust control analysis and design

    CACSD for Robust Control

    Model-based Predictive Control

    Model-based fault detection and diagnosis

    Application Fields:

    Process Engineering Systems

    Chemical/Petrochemical manufacturing processes

    Civil & Mechanical Engineering

    Structural Control

    Structural Health monitoring

    Robotics

    9Research Report 2006

  • 7/31/2019 Research Report 2006

    10/50

    Mobile Robots

    Biomedical applications

    Development of glucemic predictive models

    Insulin dosage-aid systems Computer Graphics

    New rendering algorithms based on MIA

    MICE website: http://mice.udg.es

    Computer Vision and Robotics (Vicorob)

    The Computer vision and robotics group develops real time computer vision and image processing systems with

    applications such as object, pattern, color and shape recognition and tracking of mobile objects. One of the mainapplication areas is the control quality of industrial products. Other fields of application are biomedicine and se-curity-related systems. The computer vision systems are also the main source of information of the robots deve-loped by the group. Some of them aim to improve the autonomy and mobility of disabled people, while others areautonomous and teleoperated vehicles with industrial and research applications. For instance, GARB is an un-derwater robot developed in collaboration with the UPC. GARB is used to take long underwater observationsand to collect samples in a teleoperated way. The autonomous and teleoperated robots have transport, industrialand security-related applications.

    The following are VICOROBs research laboratories:

    Underwater Robotics Lab Underwater Vision Lab

    Real-Time Hardware Lab

    3D Perception Lab

    Image Analysis Lab

    GGG website: http://vicorob.udg.es

    10 Research Report 2006

  • 7/31/2019 Research Report 2006

    11/50

    IST-2-004363 GameTools

    Project Type: STREP of IST-6FPDuration: 33 MonthsStart: September 2004Volume: 59 person-month (420 Total)Budget: 1.649.863 (245.100 UdG)Funding: 100% from IST

    Partners:

    University of Girona (UdG) - SpainUniversity Jaume I (UJI) - SpainVienna Institute of Technology (VUT) - AustriaBudapest University of Technology and Economics (BUTE)- HungaryTechnical University of Valencia (UPV) - SpainUniversite de Limoges (Unilim) - France

    Asociacin de investigacin de la industria del juguete, co-nexas y afines (AIJU) - SpainPGM Trading - AustriaSektor4 - DenmarkDigital Legends Entertainment - SpainInfowerk Softwareentwicklungs - AustriaGedas Iberia - Spain

    Coordinator:Mateu Sbert (UdG)

    Project Management Office:

    Jordi Palau (UdG) and Denisa Gibovic (UdG)

    UdG Team:Mateu SbertMiquel FeixasFrancesc Castro

    lex MndezRoel Martnez

    Lszl NeumannJaume RigauJordi PalauFrancesc GonzlezIsmael GarcaNicolau Sunyer

    Advanced Tools for Developing Highly Realistic Computer Games (GameTools)

    GameTools is a strategic project funded by the European Commission coordinated by the University of Girona. It will

    create next generation real time 3D libraries for Geometry, Visibility and Global Illumination for the PC platform using

    the OGRE 3D engine, with an extension to consoles PS2, Xbox, PS3, Xbox 360 through the use of Shark3D engine

    4. Public Funded Projects4. Public Funded Projects

    11Research Report 2006

  • 7/31/2019 Research Report 2006

    12/50

    .Geometry

    The GTP GeometryLib supplies a Multiresolution LOD (Level of Detail)Solution which addresses the shortcomings of previous approaches by

    giving developers access to a complete package of solid technologywhich supplies

    A fast image based simplification algorithm to efficiently cre-ate high quality MLODs without human intervention.

    Multiresolution triangle strip generation taking into accounttopology and texture of the base model.

    A memory & runtime efficient compressed multiresolutionmesh format that contains triangle strip information, avoi-ding the need for costly on-the-fly tristrip generation.

    A new multiresolution model specially fitted for plants allowing for an incredible amount of close up detail for e.g.

    leaves, while at the same time supplying simplified representations for the fast rendering of objects which arefurther away.

    Visibility

    In computer graphics, Visibility deals with the problem of rendering faster by rendering only the objects of the scene that can beseen.Existing solutions along that line employed in modern day 3D/game-engines are mostly based either on Portals or on Quakestyle PVS (potentially visible set). While these approaches have been successfully employed in commercial products for years,they nonetheless have considerable drawbacksThe GTP VisibilityLib overcomes these problems with a 2-phase strategy:

    A solution which supplies precalculated visibility based on modern visibility research, also suited for outdoorscenes.

    A solution which efficiently employs modern day graphic hardware to deliver on-the-fly visibility with minimaloverhead.

    Illumination

    Global Illumination is the field of computer graphics that deals with physically correct illumination. Usually this is associated withstochastic raytracing solutions taking hours to calculate a single picture.The GTP IlluminationLib brings physically correct rendering to the domain of re-altime graphics.

    Effects that will be possible include:

    Soft Shadow/Lighteffects

    Dynamic Lightsources

    Indirect Illumination

    Reflections

    Realtime Radiosity

    High-Quality Materials (Metal,...)

    Cloud Rendering

    allowing for previously unseen levels of realism in 3D/game applications.

    Research Report 200612

  • 7/31/2019 Research Report 2006

    13/50

    Research Report 2006 13

    MRTN-CT-2004-505026 - Momarnet

    Project Type: Marie Curie RTN

    Duration: 48 MonthsStart: September 2004Budget: 2.618.044 (178.421 UdG)Funding: 100%

    Partners:

    Institute de Physique du Globe de Paris - FranceInstituto do Mar, Univ. Aores - PortugalUniversidade de Lisboa - PortugalStockholm University - SwedenMax Plank Institute for Marine Microbiology - GermanyInstitute Nazionale di Geofisica e Vulcanologia - ItalyNational Oceanography Centre - United KingdomCardiff University - United KingdomUniversity of Whales - United KingdomUniversity of Girona - SpainVrije Universiteit Brussels - BelgiumU. de Bretagne Occidentale - FranceU. Pierre et Marie Curie - FranceInstitut Franais de Recherche pour lExploration dela Mer - France

    Coordinator:

    Mathilde Cannat (IPGP)

    Project Manager:Rafael Garca (UdG)

    UdG Team:Rafel GarcaPere RidaoOlivier Delaunoy

    MOMARNET

    The MOMARnet Marie Curie Research Training Network aims at strengthening the European approach to deepseafloor observatory science and technology, by reinforcing the links between fourteen European laboratories in

    eight countries, by training a multidiciplinary group of young researchers, and by producing new science and pre-

    paring for new seagoing cruises and seafloor experiments.

    MOMARnet has two broad research targets:

    Understanding the dynamics of slow-spreading ridge hydrothermal vents and their plumbing system(depth of seawater circulation, heat sources, feedback between volcanic, hydrothermal and seismicevents)

    Understanding deep-seafloor hydrothermal ecosystems, including their distribution, role and variety

    of organisms, life cycles, effects of changes in the hydrothermal outflow (temperature, flowrate,chemistry, etc.)

  • 7/31/2019 Research Report 2006

    14/50

    MOMARnet science will contribute to the implementation ofthe MOMAR (MOnitoring the Mid-Atlantic Ridge) interna-tional project of setting up a multidiciplinary sea-floor ob-servatory for monitoring hydrothermal vent environmentsalong the Mid-Atlantic ridge close to the Azores archipe-lago.

    The University of Girona team will develop a semiautomaticsoftware that willproduce well navi-gated mosaics ofseafloor images, and establish a data bank of processed seafloor imagesfor the MOMAR hydrothermal sites. These data will be key to manage fu-ture MOMAR-related experiments. PhD student will develop software tocombine navigation data, images of the ocean floor, seafloor bathymetryand acoustic imagery. Mosaicing techniques (existing and developed wi-

    thin this project) will be applied to build a geo-referenced image mosaic ofthe Lucky Strike hydrothermal site. This mosaicing software will be deve-loped in a flexible manner, so as to be expanded and applied to 3D-visionsystems, video imagery, and high-resolution acoustic systems used on cu-rrent ROVs and AUVs, and to other configurations that may be used in se-afloor surveys in the future

    Research Report 200614

  • 7/31/2019 Research Report 2006

    15/50

    Research Report 2006 15

    MRTN-CT-2006-036186- FREESUBNET

    Project Type: Marie Curie RTN

    Duration: 48 MonthsStart: November 2006Budget: 195.695 UdGFunding: 100%

    Partners:

    University of Southampton - UNITED KINGDOMCEA - FRANCECNR Robot - ITALYCNRS - FRANCECybernetix FRANCEHellenice Centre for Maritime reserach - GREECEHeriot Watt University - UNITED KINGDOMIFREMER - FRANCEIST - PORTUGALJRC - ITALYNTNU - NORWAYUniversity of Hannover - GERMANYUniversity of Craiova - ROMANIAUniversity de Girona - SPAINUniversity of Azores - PORTUGAL

    Coordinator:

    Prof. Philip Alan Wilson (U. of Southampton)

    Project Manager UdG:Pere Ridao (UdG)

    UdG Team:Pere RidaoRafel Garca

    FREESUBNET - A European research network on key technologies for intervention autonomous

    underwater vehicles

    There are many unresolved scientific questions and equally many major strategic economic and social

    development needs from an exploration of the seas around Europe. There is a pressing need for the

    development of advanced technologies to explore and exploit in a consistent and better manner the

    vastness of the oceans of the world.

    Intervention Autonomous Underwater vehicles (I-AUV) have shown their potential for the acquisition

    of marine data at extremely large scales and for the direct intervention in underwater structures without

    the need for constant human interaction and supervision. Part of the technology required for their de-

    velopment has proven to be feasible. An increased effort is now needed to further the advances already

    made in the technology and to make the results more accessible to a wider range of commercial and

    scientific end users. To this end it is necessary to create a critical mass of young researchers capableof bridging the gap between marine technology and science.

  • 7/31/2019 Research Report 2006

    16/50

    To achieve this FREESUBnet will first explore a set of applications with intervention requirements thus

    allowing the focus of the research to be in the fields of Marine Science, Inspection of Wrecks & Ar-

    chaeology, assessment of Energy systems and multi-disciplinary applications in Marine Science. In

    this context, the network will study novel concepts of IAUV' leading to the identification and the deve-

    lopment of the technologies needed to fulfil these tasks. The techniques for navigation, autonomous

    underwater telemanipulation, vehicle control and mission management technologies, will be furtherexplored and developed to close the gap between state-of-the art IAUV technology and its applications.

    FREESUBnet aims to meet its goals by establishing an international network of technology developers

    as well as being inter-sectorial and deliver the technology to end-users from different and diverse dis-

    ciplines.

    16 Research Report 2006

  • 7/31/2019 Research Report 2006

    17/50

    IST-5-012990 - ONE

    Project Type: STREP of IST-6FP

    Duration: 30 MonthsStart: September 2006Budget: 2.026.724 (154.775 UdG)Budget: 27 person-monthFunding: 100% from IST

    Partners:

    Create-NET- ItalySoluta.Net Est Europa Srl - RomaniaIstituto Trentino di Cultura - ItalyWaterford Institute of Technology - IrelandUniversitat St. Gallen - SwitzerlandService Management International Limited - UKUniversitat de Girona - SpainCoopservice S. Coop. P.A. - Italy

    Coordinator:

    Prof. Imrich Chalamatac (Create-NET)

    Project Manager UdG:

    Josep Lluis de la Rosa (UdG)

    UdG Team:Josep Lluis de la RosaMiquel Montaner

    Patricia Gaona

    ONE - Open Negotiation Environment

    The ONE project is a STREP project funded by the European Commission under the DG-INFSO that is geared

    towards SMEs in order to provide them with a trusted, secure and free of charge technological environment in

    which they are able to create tactical and strategic alliances; for the ultimate goal of pursuing business opportu-

    nities and growth.

    The project allows organizations to create contract agreements in order to supply complex, integrated services

    as a virtual organization (or coalition).

    To keep up with the competition in digital markets SMEs have the need of developing alliances and collaborating

    together in order to provide joint services and also to address large tenders.

    ONE is structured into a coherent set of activities with a balanced mix of:

    Research on economic, societal and user aspects

    Research and development of enabling technologies and methods (driven by the above)

    Research and development of enabling technologies and methods (driven by the sdfasdf

    Trials and evaluation activities.

    17Research Report 2006

  • 7/31/2019 Research Report 2006

    18/50

    ONE will be achieved through a set of technology components which can be tailored to the needs of

    a specific organization; conforming to its business requirements and adapting to its every requirement.

    The project will deliver an open source platform for ne-

    gotiation, driven by innovative business models andtechnologies; it will be implemented mainly as an inte-

    gration of innovative research component, delivered wi-

    thin the project, and existing with off-the-shelf

    technology using already standardized information mo-

    dels.

    It will extend and leverage a more general purpose di-

    gital ecosystem architecture. There will be a general ar-

    chitectural framework developed for negotiation,

    allowing the adaptation of the platform to many different

    industry sectors.

    Meta-modeling (through adoption of Model Driven Architecture ) is the approach that will be followed

    to achieve the desired high level of abstraction, reuse and code generation. While the P2P architecture

    will be based on existing open source technology, ONE will do produce innovative technologies in all

    the layers above.

    ONE will provide a negotiation mechanism that is reusable in different application and service domains

    providing a great level of flexibility and allowing users for example to create negotiation processes

    also for other goals like for example qualification or certification processes.

    The ONE platform will actively support the human negotiators by exploiting automatic learning techni-

    ques applied to the goal of learning the best negotiation strategies in a multi-agent environment. Op-

    timized strategy would be suggested through a learning process based on the observation of user

    behavior, other partners actions as perceived by the user and concurrent running negotiations.

    ONE supports a model of collaboration and trust based on the idea of "collaborative multi-agent

    systems", where agents can work and learn with other trusted agents and develop collaborative lear-

    ning schemes. We will develop models of information and knowledge sharing, taking into account pri-

    vacy concerns, based on a model of multiple agent clusters, characterized by different levels of trust.

    The policy learning algorithms will be based on the observation and storage of real interactions bet-ween the user and the negotiation service and between the user and other players (Trust), as these

    are performed in the Negotiation Execution Environment.

    The ONE negotiation environment will support technical and non-technical users in specifying strate-

    gies and rules for negotiating terms of service delivery. It will provide a distributed and peer-to-peer

    platform for fully automated execution of negotiations based on knowledge sharing and self-learning.

    18 Research Report 2006

  • 7/31/2019 Research Report 2006

    19/50

    DPI2006-09370

    Project Type: MEC

    Duration: 36 MonthsStart: October 2006Budget: 72.600Funding: 100%

    Partners:

    University of Girona

    Coordinator:

    Joaquim Melndez

    UdG Team:Joaquim MelndezJosep XargaySergio HerraizCarles PousJosep DellonderBianca Innocenti

    Fault Diagnosis of Electrical Distribution Networks Based on Cases and Models

    The project proposes a case based reasoning (CBR) strategy to diagnose electric faults. Cases will be built from disturbances,collected by power quality monitoring instruments (events) located at substations, and register reporting all the action performed

    in the network gathered in the network control center.

    Setting dependences between waveform attributes and its causes will be used in the definition of cases. Disturbance propagationmodels and feature extraction strategies will be analysed in order to represent cases consistently. This analysis will be alsouseful in the definition of Retrieval, Reuse, Revision and Retain tasks needed to build the CBR cycle.

    The Case Based Diagnosis system will be impro-ved in the location step by the integration of a mo-dule to estimate the distance to faults in thedistribution network. This module will be based inthe estimation of the impedance during the fault

    and the use of algorithmic techniques, involving themodel of the line, to estimate an equivalent fault dis-tance. The combination of this estimated distanceand origin from retrieved cases in presence of anew disturbance will enhance the diagnostic.

    Finally, an extension of cases to sequences of themis proposed to predict catastrophic faults. The re-cognition of a disturbance as part of a one of thesesequences will be interpreted as a symptom to pre-dict a possible fault.

    19Research Report 2006

  • 7/31/2019 Research Report 2006

    20/50

    DPI2006-15476-C02-02

    Project Type: MEC

    Duration: 36 MonthsStart: October 2006Budget: 86.394Funding: 100%

    Partners:

    University of Girona

    Project Manager UdG:

    Joaquim Armengol

    UdG Team:Joaquim ArmengolRemei CalmLuis Eduardo MjicaEsteban GelsoMaira Alejandra Garca

    AUTOMATED FAULT DETECTION, DIAGNOSIS AND TOLERANCE IN UNCERTAIN AND DISTRIBUTED

    SYSTEMS

    Fault detection, diagnosis and tolerance in systems allow maintaining the process in a high level of production and reliability. Thepresent techniques show problems when they are applied to uncertain and distributed systems. For this reason, they are beingresearched in a very active way to develope new techniques.The most important objectives of this project are to decrease thehigh false positive and false negative rates, to improve the computacional efficency of fault detection and diagnosis and to au-tomate the identification of components that are the cause of the failures.

    These problems for the uncertain systems can be treated using the classical diagnosis approaches: the FDI (Fault Detectionand Isolation) community, whose foundations are based on engineering disciplines, such as control theory and statistical decisionmaking, and the DX (Diagnosis) community, whose foundations are derived from the fields of Computer Science and ArtificialIntelligence and bridging the methodologies of the FDI and DX Communities (Bridge). Firstly the project proposes this last ap-proach to automate fault diagnosis in these systems. The Research Group in Girona has developed the SQualTrack tool forthe fault detection of these systems and this project proposes its extension for diagnosis with techniques derived from modal in-terval analysis. We consider that it is important the coordination with the Quivir (Sevilla and Huelva University) Group that is also

    studying these same systems using constraint satisfaction techniques, guaranteed methods and data-mining techniques. Es-pecially, the last techniques can work without the knowledge of the model and can deal with the empiric data (data-based diag-nosis). These techniques are being validated in real systems.

    In distributed systems, firstly this project addresses the general problems associated with automatic determination of the allocationof sensors to improve the diagnosticability of the distributed systems. Secondly we will develop a framework for self-diagnosisof distributed systems where we take into account the persistency of the different models and different improvements based onclustering and distributed constraint satisfaction techniques.

    Finally, we consider that the dependability and fault tolerance are important in the systems. So, we propose the modelling of re-aliability and security properties and the identification of the faults in the monitoring systems. We will develop different algorithms

    that take this information and automatically obtain the fault diagnosis and choose recovery actions using planning and re-planning. It can hopefully provide a foundation for building realistic automatic dependability and tolerance systems.

    20 Research Report 2006

  • 7/31/2019 Research Report 2006

    21/50

    TEC2006-03883

    Project Type: MEC

    Duration: 36 MonthsStart: October 2006Budget: 90.750Funding: 100%

    Partners:

    University of GironaUniversity of ExtremaduraUniversidad del Norte (Colombia)University of LondonTechnical University of Denmark

    Coordinator:

    Josep Lluis Marzo

    UdG Team:Josep Lluis MarzoTeodor JovPere VilLluis Fbrega

    Antonio BuenoEusebi CalleJoan PuggalSnia Oliveras

    Anna Urra

    Luis Fernando CaroFernando SolanoJeimy Vlez

    Multilayer Mechanisms for Resource allocation and Restorable Routing in GMPLS-based core networks

    (M2R3)

    nternet Protocol (IP) networks currently operate over a semi-static transport infrastructure, where the management is concen-

    trated within the IP layer. With the introduction of Multiprotocol Label Switching (MPLS), an intermediate circuit-switching com-

    munication layer is added together with its corresponding management plane. Moreover, the development of optical technologies

    such as Dense-Wave Division Multiplexing (D-WDM), supporting wavelength switching, etc., allow new multilayer networks tobe constructed with an unprecedented degree of freedom and flexibility.

    Nevertheless, the management of such multilayer networks is more complex and presents potential instabilities. The complexity

    of the evaluation of the percentage that should be applied at each management layer, i.e. IP, MPLS or optical, presents a sig-

    nificant challenge. We denote as activation degree such percentage which will depend on multiple conditions such as: the to-

    pology of the network, types of traffic, the load of the network and the desired degree of protection.

    The main objectives of the M2R3 project are the investigation of the following components: quality of service routing mechanisms,

    routing with protection and a dynamic resource management. They must be developed and configured according to the required

    activation degree. The project will provide novel tools that will enable network operators to suitably manage their networks in a

    holistic manner, allowing the various technologies to operate in concert rather than inefficiently or even in conflict with each other.

    To the best of our knowledge, the approach we present to resolve this complex management challenge has not been undertaken

    in other proposals.

    Research Report 2006 21Research Report 2005

  • 7/31/2019 Research Report 2006

    22/50

    CIT-020600-2005-39

    Project Type: PROFIT

    Duration: 12 MonthsStart: July 2005Budget: 122.984 (32.067 UdG)Funding: 47.600 (16.000 UdG)

    Partners:

    Plastiasite S.A.University of GironaFundacin ASCAMMEleda-CumsaSogefi Filtration S.A.

    Coordinator:

    Ricard Jimnez (Plastiasite)

    Project Manager UdG:

    Joaquim Melndez

    UdG Team:Joaquim MelndezJoan Colomer

    NetMould. Efficient technology integration of the mould in knowledge based networks

    This projects aim was to elaborate a viability study of a system that retains the know-how of a traditional industrial sector in ahigh technology level such as plastic injection companies (mould builders, parts designers or plastic injection plants), which isgoing through difficulties due to the relocation process.Netmould demonstrated that it is feasible to develop a solution to provide the sector with a technology that has an added valuedirectly incorporable to moulds and its environment(the plant or laboratory). The key is to shift the view-point to a mould-centric vision that transforms thecurrently passive tools into network enabled ele-ments with intelligent knowledge embedded.The results of the project show that this emerging

    technology assures a competitive advantage (re-duction in production time and errors), a quality en-hancement (thanks to preventive diagnosisenhanced by sharing the information on-line) andan increase in usability (process monitoring, remotecontrol).The tasks of the UdG team have been oriented toimprove the diagnosticability of moulds using statis-tical techniques. Multiway Principal ComponentAnalysis has been used to model injection of plasticpieces and those models have been exploited todetect and diagnose abnormalities in future injec-tions.

    22 Research Report 2006

  • 7/31/2019 Research Report 2006

    23/50

    Research Report 2006

    PTR1995-1020-OP

    Project Type: MEC

    Duration: 12 MonthsStart: July 2006Budget: 57.380 Funding: 100%

    Partners:

    University of GironaEndesa

    UdG Team:JoaquimMelndezSergio HerraizCarles Pous

    Development of a prototype to assist fault location in distribution networks

    This project aims at developing a prototype tool to assist fault lo-

    cation in distribution networks using the companys existent kno-

    wledge about network and perturbation logs (wave shapes and

    events) as well as operations and actuations on the network re-

    gistered in the control center.

    Two main aspects in the monitorization of the power quality areexpected to be improved:

    Analysis of disturbances propagation, conciliating theincidence logs (from network control center) with thedisturbances registered in different locations of the net-work (using quality registers)

    Location of faults occurred in the distribution. Asystematic analysis of waveforms of disturbances using

    algorithmic methods of fault location will allow reducingcurrent location times.

    The final purpose is to characterize and systematically eva-

    luate perturbations in the distribution network using previous

    logs and its reutilization to perform a fast location and diagno-

    sis of the network distribution faults.

    Fault location consists in finding an accurate estimation of the

    zone in which the fault has occurred, using the measures of

    the waveform of the registered event and the network topo-

    logy.

    23Research Report 2005

  • 7/31/2019 Research Report 2006

    24/50

    XARTAP

    Project Type: Generalitat de Catalunya

    Duration: 24 MonthsStart: January 2006Budget: 37.724.Funding: 100%

    Partners:

    Centre CIMCentre Internacional de Mtodes Numrics en EnginyeriaUniversitat Politcnica de CatalunyaCentre Visi per ComputadorUniversitat de GironaConsell Superior dInvestigacions Cientfiques

    Coordinator:Centre CIM

    Project Manager UdG:Xavier Cuf

    UdG Team:Marc CarrerasJavi CobosFranois ChungJordi FerrerJordi Freixenet

    Josep ForestRafael GarciaXavier LladJoan MartRobert MartMarta PeracaulaPere RidaoJoaquim Salvi.

    Reference network in Advanced Production Techniques

    The main objective of the task carried out by VICOROB Research Group in this project is developing a set of prototypes of Iden-tification and Positioning Systems.

    The final result of the project will be the development of different solutions that allow identifying the different parts that constitute

    an industrial kit, and obtaining the position and orientation with enough accuracy to allow picking these parts using a manipulator

    or grasping system. Usually, the different parts arrive all together to the inspection point in a disordered bunch, and some parts

    can present different kinds of deficiencies.

    Recognition System: different computer vision algorithms will be developed to identify the different parts that can appear at the

    inspection point. This system will be able to extract geometric information from the images and compare it to the corresponding

    CAD models in order to determine whether the part is faulty or not, or if the visual information obtained is adequate or more ad-

    ditional images are required.

    Position and Orientation of the parts: a robust computer vision algorithm will be developed to determine position and orientation

    of the previously recognized part referred to a determined reference frame.

    24 Research Report 2006

  • 7/31/2019 Research Report 2006

    25/50

    Research Report 2006

    DPI-2005-08668-C03-02 (CHIMPFA)

    Project Type: MEC

    Duration: 36 MonthsStart: December 31, 2005Budget: 71.400 Funding: 100%

    Partners:

    University of GironaTechnical University of Catalonia (CoDaLab)Technical University of Catalonia (CITCEA)

    Coordinator:

    Jose Rodellar

    Project Manager UdG:

    Ningsu Luo

    UdG Team:Ningsu LuoJoaquim ArmengolIns FerrerMiquel CompteRodolfo VillamizarSandra M. CastilloJorge E. Florez

    ANALYSIS, IDENTIFICATION AND CONTROL OF MECHATRONIC SYSTEMS WITH HYSTERESIS AND

    FRICTION. APPLICATION TO PIEZOELECTRIC AND MAGNETORHEOLOGICAL ACTUATORS

    Hysteresis and friction are nonlinearities encountered in various systems in science and engineering areas. This project dealswith mechatronic systems equipped with actuators based on smart materials, in particular piezoelectric and magnetorheological.These actuators are particularly useful in high performance control applications, due to their high precision and fast response.These advantages can only be exploited through a fine constructive process and if appropriate theoretical tools are available.Indeed, these actuators have a complex dynamics with presence of nonlinearities, hysteresis and friction being the most signi-ficant. Due to the high performance required in closed loop, these actuators are prone to the uncertainties in the models and inthe operation conditions. Thus it is essential to have appropriate analytical tools for the study and robust identification of their dyna-mic models. Moreover, the design of controllers needs the use of advanced nonlinear control theories to be able to guaranteerigorously the narrow specifications in terms of stability, precision and robustness.This project follows three complementary research directions:

    Development of new theoretical tools for dynamic analysis, robust identification and control of systems withhysteresis and friction;

    Construction of prototypes of piezoelectric and magnetorheological actuators integrated in mechatronic systems;

    Experimental validation in these systems of the developed methods and algorithms.

    The project will consider two problems: micropositioning control and vibration reduction, which have special practical interest no-wadays in areas as automotive industry, mechanics, aeronautics, aerospace and civil engineering, as well as in nanotechnolo-

    gies.

    25Research Report 2005

  • 7/31/2019 Research Report 2006

    26/50

    DPI2005-09001-C03-01 (Airsub)

    Project Type: MCYTDuration: 36 MonthsStart: December 31, 2005Budget: 158.000 Funding: 100%

    Partners:

    University of GironaTechnical University of Catalonia (UPC)Universitat de les Illes Balears

    Coordinator:

    Joan BatlleProject Manager UdG:

    Pere Ridao

    UdG Team:Joan BatllePere RidaoMarc CarrerasDavid Ribas

    Andres El-FakdiElisabet BatlleNarcs Falgueras

    Josep CortadaJos Luis MuozFerran GironesElibol ArmaganAntonio Tiano

    AIRSUB - Underwater Robotics Industrial Applications

    This project proposes the design and development of the necessary

    technologies for the adaptation of the prototypes of the low-cost un-

    derwater vehicles GARBI-ROV, GARBI-AUV , URIS and RAO II fortheir use in shallow waters. First, in collaboration with industry, a ca-

    talogue of industrial applications -feasible for low-cost underwater

    vehicles- will be created. For each and every of these missions, the

    adequate scenario will be established, as well as a mission prelimi-

    nary schedule. Next, the project will be focused in 3 industrial appli-

    cations:

    Dam inspection of hydroelectric plants

    Underwater cable and pipe inspection

    Inspection of harbours.

    26 Research Report 2006

  • 7/31/2019 Research Report 2006

    27/50

    Research Report 2006

    Every one of these applications will be leaded by one of the 3 subprojects, will be associated to a market-leader

    company in its field, and will end with a practical demonstration.

    For dam inspection we propose the use of an AUV which could

    scan the dam wall at the same time it acquires a video, registe-

    ring its state. This video will be processed off-line to build a ge-oreferenced photo-mosaic, which will allow the company

    engineers to gain a global view of the condition of the dam.

    Analysing this mosaic, successive inspections can be scheduled

    to evaluate selected areas with higher resolution, as well as

    studying its temporal evolution. This inspection technique will be

    applied both to both the grilles of intake and the penstock gallery.

    Cable and pipe inspection will be carried out using an AUV that

    will first submerge and detect the cable. Then, the robot will be

    guided by means of the automatic visual tracking system. Along

    the mission, the AUV will acquire, in a first phase of the project, a large set of images that will be processed off-

    line (building of mosaics, analysis of an expert, etc.). In a second phase the AUV will filter the acquired images,keeping those images which show defects of the cable/pipe and annotating their localization. In this way, the com-

    pany engineers can concentrate in the suspicious areas along the inspected section of the cable/pipe. Finally, for

    the inspection of the breakwaters walls of harbours a ROV will be used. This robot will acquire images which will

    be used to build a georeferenced mosaic. Fuzzy techniques will be used to localize faults and anomalies in the

    moles, such as their temporal degradation.

    27Research Report 2005

  • 7/31/2019 Research Report 2006

    28/50

    CERTAP - PVPC

    Project Type: CeRTAP (Generalitat de Catalunya)

    Duration: 36 MonthsStart:April 1, 2005Budget: 321.275,37 (64.920 )Funding: 100%

    Partners:

    Technical University of Catalonia (UPC)Autonomous University of Barcelona (UAB)Girona Graphics Group (University of Girona - UdG)VICOROB (University of Girona - UdG)CIM Centre

    Coordinator:

    Pere Brunet (UPC)

    Project Manager UdG (GGG):

    Gonzalo Besuievsky

    UdG Team:Xavier PueyoGonzalo BesuievskyAlbert Mas

    Virtual Prototyping for Architectural Projects

    The main goal of this project is the design and development of a system for fast virtual prototyping models in order to providehelpful tools for architects and building designers in early phase of a project. The system has the following features:

    Automatic architectural recognition symbols and vectorization of freehand 2D drawing floorplans

    Automatic generation of 3D models from 2D floorplans

    It allows edition for basic installation as light sources and furniture with texture edition

    Fast lighting visualisation for both natural and artificial light sources

    Analysis and inspection of the generated prototype in lower-cost virtual reality platforms: stereoscopic workbenchor standard computer with 3D immersive devices

    Interactive re-edition of the model openings with natural lighting visualization

    Navigation and inspection tools for remote collaborative work

    The whole system is integrated in a single application and it is designed in a way that could be integratedwithin any standard CAD architectural platform as a complement for fast virtual prototyping and validation de-signs.

    28 Research Report 2006

  • 7/31/2019 Research Report 2006

    29/50

    CERTAP - PVPC

    Project Type: CeRTAP (Generalitat de Catalunya)

    Duration: 36 MonthsStart:April 1, 2005Budget: 321.275,37 (64.920 )Funding: 100%

    Partners:

    Technical University of Catalonia (UPC)Autonomous University of Barcelona (UAB)Girona Graphics Group (University of Girona - UdG)VICOROB (University of Girona - UdG)CIM Centre

    Coordinator:

    Pere Brunet (UPC)

    Project Manager UdG (VICOROB):

    Xavier Cuf

    UdG Team:Xavier CufJoan Mart

    Arnau OliverTudor NicoseviciElisabet Batlle

    Virtual Prototyping for Architectural Projects

    The aim of the Project is to create a tool for helping the professionals of building design and architects.

    The base of the work is the experience of the different research groups that belongs to CERTAP in topics such

    us computer vision, vectorial designs generation from hand made layouts, computer assisted design, illumination

    and generation of realistic images and virtual reality. As we said before, the final objective of the project has been

    to obtain an immersive tool for supporting the professionals the design of buildings. The opinion of the professio-

    nals had been considered in all the steps of the development of the project. The main characteristics of the de-veloped tool are:

    2D sketch generation and the 3D model from hand made layouts. At this step, it is necessary tot take

    into account buildings with different levels, stairs and different kinds of ceilings.

    Realistic Illumination with ambient simulation (using radiosity), using simulation of natural and arti-

    ficial light. The application offers the possibility to visualize the digital model surrounded by images

    belonging to the reality.

    Analysis and inspection of virtual prototypes generated by the system in low cost virtual reality plat-

    forms (stereoscopic table) or using standard computers equipped with 3D immersive devices.

    Tools of remote working that allows the cooperative inspection and analysis between different teams

    29Research Report 2006

  • 7/31/2019 Research Report 2006

    30/50

    working, using stereoscopic tables, in different locations.

    The work developed by the VICOROB research group of the UdG has been:

    The architect provides the application with a hand-made sketch, and then the application generates a full 3D

    model of the building. Once the 3D model has been generated, the professional has the option to add texture and

    simulate illumination (natural and artificial), being able to obtain realistic renderings. On the one hand, the resultscan be used by the architects to observe and adjust different aspects of the building at pre-project stage and, on

    the other hand, to help the architects interact with the customers.

    This task includes a sub-application that allows recovering the texture from an object of interest that appears in a

    real scene. This real texture could be projected over virtual surfaces belonging to the model (walls, roofs, )

    using the adequate scale.

    30 Research Report 2006

  • 7/31/2019 Research Report 2006

    31/50

    TIN2004-06354-C02-02

    Project Type: MCYT

    Duration: 36 MonthsStart: December 2004Budget: 72.000 Funding: 100%

    Partners:

    University of Girona - SpainTechnical University of Valencia - Spain

    Coordinator:

    Federico Barber (UPV)

    Project Manager UdG:

    Beatriz Lpez

    UdG Team:Beatriz LpezSantiago EstevaJosep M. HumentOrlando ContrerasSilvia SurezChristian QuinteroMiquel MontanerDidac BusquetsBianca Innocenti

    Developing Distributed Computing Techniques for Scheduling. Application to Transport Problems

    This project aims at the development of distributed computing techniques and multi-agent systems for the resolution of constraintsatisfaction and optimisation problems (CSP, COP), these techniques particularly guided towards planning and scheduling pro-cesses. So that, it is a coordinated project in which two research groups work together with complementary experience on: (i)CSP techniques, distributed CSP, planning and scheduling and (ii) distributed and open frameworks, multi-agent systems andcoordination mechanisms. Furthermore, the integration of planning and scheduling processes is considered in the CSP frame-work. The scientific-technique objectives are very relevant and important allowing us to manage problem typologies in whichtraditional techniques are not appropriate. Additionally, in the research project it is sought to apply and evaluate the developed

    techniques in evaluation environments broadly contrasted in the scientific community and, mainly, with two concrete and realframeworks based on rail transport and intercity transport in which it is maintained respective agreements of collaboration.The main objectives of the project are:

    Development of scheduling techniques that integrate algorithms from operational research, topological methodsand intelligent techniques oriented towards the efficient treatment of complex scheduling problems.

    Extend the applicability of constraint satisfaction techniques with both multi-agent techniques and distributed ap-proaches of satisfaction problems.

    Development of an integrated architecture for planning and scheduling.

    Application of the architectures and techniques in real environments (train infrastructure planning, crew sche-

    duling for road transport) as well as in scientific benchmarks (Pacifica).

    31Research Report 2006

  • 7/31/2019 Research Report 2006

    32/50

    TIN2004-08065-C02-02

    Project Type: MEC

    Duration: 36 MonthsStart: December 2004Budget: 146.000 Funding: 100%

    Partners:

    University of Girona - SpainTechnical University of Catalonia - Spain

    Coordinator:

    Antoni Sellars (UdG)

    UdG Team:Joan Antoni Sellars ChivaFrancesc Xavier Pueyo SndezJordi Regincs IsernInmaculada Boada OliverasNarcs Coll ArnauMaria Ethel GuerrrieriYago Diez DonosoMarta Fort MasdevallNarcs Madern LeandroDavid Fguls MassotMiquel Bofill Arasa

    Antoni Bardera Reig

    Albert Rovira BrugusJos Suy FranchJosep Soler MasJos Antonio Mndez MndezIsabel Pelez Hernndez

    Interfaces avanzadas para la interaccin transparente con modelos altamente complejos en entornos de

    realidad virtual

    The development of effective user interfaces for virtual reality environments must face different challenges. Not

    only must one evolve new paradigms that are adequate in those environments, but they must also be minimally

    invasive (not to spoil the immersion illusion), and they must rely on an adequate support from the underlying mo-

    dels and model-handling functions, so that the maximum complexity of usable scenes is not seriously limited.

    This project aims at pushing further the state of the art, developing new techniques at the modelling level (to allow

    handling models like the most complex currently available, which are too long to fit in memory for a reasonably

    configured workstation). It also aims at perfecting the state of the art concerning the interfaces themselves, de-

    veloping specific techniques adapted to the interaction in those media, and especially for remote collaboration.

    Moreover, the results will be tested developing specific prototypes to verify the usability of the proposed techniques

    from the point of view of experts in three application areas, chosen amongst those that may benefit the most from

    the expected results.,

    32 Research Report 2006

  • 7/31/2019 Research Report 2006

    33/50

    TIN2005-08945-C06-03

    Project Type: MEC

    Duration: 36 MonthsStart: December 2005Budget: 47.600 Funding: 100%

    Partners:

    Universidad Nacional de Educacin a Distan-cia - SpainUniversity of Girona - SpainTechnical University of Valencia - SpainUniversity of Castilla la Mancha - SpainUniversity of Granada - SpainUniversity Carlos III of Madrid - Spain

    Coordinator:

    Jesus Gonzalez Boticario (UNED)

    Project Manager UdG:

    Ramon Fabregat

    UdG Team:Ramn Fabregat GesaEusebi Calle OrtegaMontse Tesouro CidDavid Victor Hugo Mrida

    Cesar Augusto Mantilla NavarroJoan Puiggali AllepuzPau Sanz HellinLiliana Mara Carrillo FlorezClara Ins Pea de CarrilloCarlos Arteaga Rivero

    Adaptation based on machine learning, user modelling and planning for complex user-oriented tasks

    (AdaptaPlan)

    This project has two main objectives. On one side, to analyze the capability of automatically solve tasks that needthe integration of planning, machine learning and interaction with differentagents (human or software) to ubiquitously and dynamically adapt to theevolving needs of the context. Thus, we have defined a set of scientific goalsrelated to user modelling, machine learning, planning, multi-agent systems,and ubiquitous computing.

    On the other side, we would like to benefit from the two-fold experience ofthe project researchers, both as users and developers, in a specific applica-tion domain: university education. In this domain, tools that integrate pre-vious techniques are strongly needed. This applied objective consists in thedesign and development of an architecture that provides services to bothstudents and lecturers in the new education tasks that are defined by the European Space for Higher Education.

    33Research Report 2006

  • 7/31/2019 Research Report 2006

    34/50

    Improvement and Application of the Game Technology in Virtual Reality and web con-

    tents

    Cutting-edge technology in the field of Computer Graphics is currently found in 3D engines for computer games.

    This is due to the successful combination of research conducted by universities with the traditional techniques used

    in the digital games sector. While this technology was first developed for computer games, it is expanding its

    scope of application to include other sectors which are less developed as far as realistic rendering is concerned.

    Virtual Reality and 3D web content are the sectors which show the greatest interest in incorporating the new tech-

    nologies. Therefore the purpose of this project is twofold: To continue the research on computer games techno-

    logies carried out in the project TIC2001-2416 and to apply the new technologies to the fields of Virtual Reality

    and 3D web content.

    The objectives of the project are: Improve videogame technologies in the fields of illumination and geometry

    Extend videogames technology to Virtual Reality and 3D Web

    Transfer scientific results to industry

    TIN2004-07451-C03-01

    Project Type: MEC

    Duration: 36 MonthsStart: December 2004Budget: 131.790 Funding: 100%

    Partners:

    University of Girona - SpainUniversity Jaume I - Spain

    AIJU

    Coordinator:

    Mateu Sbert (UdG)

    UdG Team:Mateu SbertMiquel FeixasJaume RigauEsteban del AceboFrancesc CastroRoel MartnezPere Pau Vazquez

    Alex MndezJordi Cat

    34 Research Report 2006

  • 7/31/2019 Research Report 2006

    35/50

  • 7/31/2019 Research Report 2006

    36/50

    DPI2004-07167-C02-02

    Project Type: MCYT

    Duration: 36 MonthsStart: December 2004Budget: 85.000 Funding: 100%

    Partners:

    Technical University of Valencia - SpainUniversity of Girona - Spain

    Coordinator:

    Jorge Bonda (UPV)

    Project Manager UdG:

    Josep Veh (UdG)

    UdG Team:Miguel Angel Sainz

    Arcadi GuitartPau HerreroRal FernndezJian WanGabriela CaldernLambert Jorba

    Development of a glucemic prediction and insulin dosage-aid system for patients with

    diabetes mellitus I (INSULAID)

    Diabetes mellitus type I is a chronic disease of great individual and social impact which affects a great number of

    people. It produces important damages in organs like eyes, kidney, heart, etc. leading to blindness, renal insuffi-

    ciency, heart attack, etc, as well as a significant reduction in life expectancy. These facts are directly related to the

    glucose levels in blood. The glucemic control in normality levels is the main objective of the treatment for diabetes

    mellitus I.

    The development of glucemic predictive models will allow a better metabolic control and, thus, reduce the com-plications of the disease. The metabolic model of the glucose due to Cobelli is nowadays widely validated in clinic

    experiences. This model is non-linear and with more than 30 parameters to identify. The objectives of this project

    are the parametric identification of this model for particular patients and the development of a training and simu-

    lation system for guaranteed prediction of patients glucemia. Based on it, an insulin dosage-aid system will also

    be developed. With these two systems the quality of life and life expectancy of the patient will be increased. For

    it, techniques for the resolution of constraint satisfaction problems in continuous domains will be used. These

    ones have experimented great advances due to the interval methods and combination with programming and op-

    timisation techniques.

    36 Research Report 2006

  • 7/31/2019 Research Report 2006

    37/50

    DPI2005-08922-C02-02

    Project Type: MCYT

    Duration: 36 MonthsStart: December 2005Budget: 76.400 Funding: 100%

    Partners:

    University of Girona (Lequia Group) - SpainUniversity of Girona (ExIT Group) - Spain

    Coordinator:

    Maria Dolors Balaguer (Lequia)

    Project Manager UdG:

    Joan Colomer (ExIT)

    UdG Team:Carles PousDaniel Alexandre MacayaFco. Ignacio GameroMontse RubioMagda Liliana Ruiz

    Development of an Intelligent Control System applied to a Sequencing Batch Reactor

    (SBR) for the removal of Organic Matter, Nitrogen and Phosphorous

    .This project is focused on the improvement of the control

    system applied to sequenced batch reactors developed in the

    project DPI2002-04579. The promising results obtained in the

    previous project allow to consider the development of a more

    ambitious control system to improve the overall process per-

    formance and adapt it according to the influent water charac-

    teristics.On the one hand, the research into the operating conditions will

    be extended to cover the simultaneous elimination of Organic

    Matter, Nitrogen and Phosphorous. In this way, the necessary

    knowledge about the relationship between trends of on-line

    variables and changes in the process will be obtained. Then,

    from this knowledge, a control system to optimize the duration

    of each stage will be implemented by means of an on-line de-

    tection of critical points.

    On the other hand, an Intelligent Control Module will be developed and implemented to diagnose the evolution of each batch

    and to optimize the schedule in order to minimize the effect of the influent water variability.

    37Research Report 2006

  • 7/31/2019 Research Report 2006

    38/50

    Development of a Modular System to Build Georeferenced Photo-mosaics of the Ocean

    Floor

    In this project we propose the development new algorithms and techniques for the automatic construction of ocean floor photo-mosaics. Photo-mosaics allow the geological analysis of ocean floor morphology and hydrothermalism of selected areas. Like-wise, underwater photo-mosaics provide high quality images of specific areas for the study of biological communities.

    The works which have been carried out up to now in the field of auto-matic construction of photo-mosaics have been limited to processingplanar environments, while most of the regions of interest for the scien-tific community are located in areas with 3-dimentional relief. For thisreason, research and development of new algorithms is a must to en-able simultaneous photo-mosaicking and estimation of underwater 3Drelief. On the other hand, none of systems described in the literature isable to automatically build large-scale photo-mosaics, due to the accu-mulated drift as the mosaic increases its size. In this project we proposeto develop new sensor fusion strategies to take advantage of all the in-formation available from the on-board sensors to automatically buildbigger and more reliable photo-mosaics.Finally, the proposed system should be modular and flexible, to adaptto a wide diversity of underwater vehicles equipped with various types

    of sensors. Such a system will allow the analysis of the temporal evolution of both existing ecosystems and morphology of the

    underwater terrain.

    CTM2004-04205

    Project Type: MEC

    Duration: 36 MonthsStart: December 2004Budget: 166.550 Funding: 100%

    Partners:

    University of Girona - SpainInstitut de Physique du Globe CNRS - FranceUniversity of Miami - USA

    Coordinator:

    Rafael Garcia (UdG)

    UdG Team:Rafael GarciaXavier CufLluis PachecoFerran ChicJordi FerrerMart Fbregas

    Antonio MartoranoTudor NicoseviciCarles MataboschXavier Muoz

    38 Research Report 2006

  • 7/31/2019 Research Report 2006

    39/50

    2003/XT/0037 R-GMPLS

    Project Type: Thematic Network

    Funding: Generalitat de Catalunya

    Partners:

    University of Girona - SpainCTTC - SpainUPC- CCABA - SpainUPC - TSC - SpainUniversidad de Mlaga - SpainUniversidad de Extremadura - SpainUPC - Telemtica - SpainUniv. Castilla la Mancha - SpainUPM - Spain

    Coordinator:

    Jose Luis Marzo (UdG)

    UdG Team:Teodor JovRamon FabregatPere VilEusebi Calle

    Management of circuit-switched IP networks (R-MPLS)

    The Thematic Network Management of circuit-switched IP networks, partially funded by the Catalan Government, started onJanuary 1 2003. The aim of this project is to integrate research on GMPLS/MPLS done in Spanish research organizations. Theproject organizes the Workship MPLS networks annually.

    39Research Report 2006

  • 7/31/2019 Research Report 2006

    40/50

    5. Technology Transfer projects5. Technology Transfer projects

    Endesa

    Project Type: Technology TransferDuration: 12 MonthsStart: February 2005Budget: 21.810

    Partners:

    Endesa Distribucin Electrica SA

    Coordinator:

    Dr. Joaquim Melndez

    UdG Team:Joaquim MelndezJoan ColomerFrancisco GameroSergio Herraiz

    Electrical fault localisation and Disturbances Propagation Analysis

    Electrical fault localisation

    Objective: The objective of this project is to delimit the location where fault has occurred by means of analysingthe wave shape of the event registered in the substation and using the additional information about the distributionnetwork obtained from the company.Tasks:

    Elaboration of an updated state-of-the-art about the algorithmic methods currently used to locatefaults using single end measurements in substations.

    Implementation and test of the methods using real data. Tuning of the modules and final validation.

    Development of an executable prototype to be used in a PC environment on the substation.

    Disturbances propagation analysis

    Objective: The objective here is to extract knowledge from the voltage sags and to associate their characteristics(shape, length, depth) to the information from the network control centre (MV/HV origin, protections, causes, han-dling, etc.) in order to diagnose similar situations.Tasks:

    Development of an automatic tool to conciliate the re-gistered incidences in the control centre and the per-turbations registered in electrical substations.

    Exploitation of the conciliated information in order toextract knowledge about causes and origins of thevoltage sags: correlation event/incidences. Determi-nation of the MV/HV characteristics and their relationto their location and the different types of protectionusing the characteristics of the events.

    40 Research Report 2006

  • 7/31/2019 Research Report 2006

    41/50

    Optimus

    Project Type: Technology Transfer

    Duration: 9 MonthsStart: July 2005Budget: 13.320

    Partners:

    Optimus

    Coordinator:

    Pere Ridao

    UdG Team:Pere Ridao

    Andres El-FakdiNatalia HurtsJoan MassichNarcs Palomeras

    Implementation of a method to detect faults in loudspeaker lines

    The main objective of this contract has been the study and implementation of a method to detect faults in louds-peaker lines. This method has been based in the addition of certain frequencies to the reproduced sound so thatthey can be recovered a posteriori. This way it is possible to identify problems within the line.In the first stage of the project we have carried out a theoretic analysis of the viability of such method using the

    signal processing tools included in Matlab.In the second stage we have used a commercial DSP together with its programming environment in order to eva-luate its viability to face the problem. We have designed and build the hardware modules needed by the system,developing a module for analogical-digital conversion and a module for input/output that have been both integratedin the evaluation board.Finally we have implemented different software modules to run the application, that perform different tasks suchas generate, send, recover and filtrate signals, as well as calculations to determine variations in line load.

    41Research Report 2006

  • 7/31/2019 Research Report 2006

    42/50

    6. Education6. Education

    Masters Programs

    Master in Automatic Control and Computer Engineering

    The Masters in Automatic Control and Computer Engineering aims to respond to the growing demand in this area

    for specialists capable of dealing with and participating in Research, Development and Innovation (R+D+I) projects

    in the production and service sectors.

    This masters programme offers state of the art training in the following areas: computer vision, robotics, control

    engineering, communications and intelligent systems. Training in these areas leads to the acquisition of techniques

    and methods and their application to solve current problems in applied fields such as medical image interpretation,

    efficient data retrieval in process supervision, quality control through artificial vision, applying AI techniques to de-

    cision making, the creation and management of distributed applications, communication network management,

    etc.

    This masters program has been selected by the Catalan and Spanish Governments to be official and conse-

    quently it is included in the official catalogue of both governments to start next September 2006. This master

    offers two specialisation branch; one in control Engineering and Intelligent Systems (ECSI) and another in Com-

    puter Vision and Robotics (VCR) and allows students following their career at the doctorate program. In the second

    academic year, master students will have the opportunity to make their stay at other universities or research cen-

    tres receiving the highly qualified personal supervision of their research and/or to get the traineeships in the

    leading industrial companies collaborating with the master program.

    More information at http://eia.udg.es/master-imagis

    VIBOT - Erasmus Mundus Masters in VIsion and roBOTics

    The European master on Computer Vision and Robotics (VIBOT) meets the present needs of in-

    dustry for quality control and automation of industrial processes as well as in the field of health

    with the ever increasing importance of medical imagery in all its forms.The key techniques are

    computer vision and robotics.

    Current research in all these domains is today particularly dynamic and the results obtained by the teams of re-

    searchers worldwide find fast outlets in industry, where the needs are increasingly important. This is relevant to

    a wide range of sectors, as for example the car industry, the agro-alimentary field or the domain of health, withthe active development of e-Health solutions.

    The continued evolution of this research sector requires adapted and extremely specialized formations. Through

    strong links in established research collaborations, our respective laboratories decided to combine their interna-

    tional experience and expertise in order to propose a new high level, international study programme running over

    four semesters. The formation intends to be at excellence level. It will benefit from the complementarities and

    synergies of our laboratories.

    Starting from a comprehensive coverage of the perequisites in the field of digital imagery and basic image pro-

    cessing, the students will cover at specialized level all subjects of artificial vision applied to robotics and medical

    imagery, as well 3D vision.

    42 Research Report 2006

  • 7/31/2019 Research Report 2006

    43/50

    This Masters programme will provide, through the proposed European formation, an unrivalled knowledge in com-

    puter vision and robotics applied to industrial problems and medical imagery. Moreover, the students will have also

    acquired a mobility, a broadmindedness and the language skills which constitute, in an age of globalization, a

    major asset for finding research work either in a public laboratory or in a private company.

    The VIBOT is a collaboration between the Universit de Bourgogne (Le Creusot, France), the Universitat de Girona (Girona,

    Spain) and Heriot-Watt University (Edinburgh, Scotland).

    More information at http://www.vibot.org

    Master in Computing

    Information Technology is one the major driving forces of modern civilization. The professionals in many areas of science and

    technology are expected to be proficient in various disciplines of the information technologies, thus guaranteeing efficiency and

    confidence in their work.

    The Master program aims at providing a solid background in Computing for qualified professionals that must able take high res-

    ponsability positions in industry or start research activities in theoretical or practical aspects of Computing. The Master also offers

    the opportunity to focus on one of its specializations:

    Algorithmics and Programming

    Information Systems

    Visualization, Virtual Reality, and Graphic Interaction

    The student can select between achieving a generic education on several subjects or focus on one of the specializations.

    The students completing the master will be competent for:

    Professional jobs that take strategic decisions, requiring a high capacity of analysis and the resolution of complex

    problems

    Creating and transferring research-based knowledge in the professional world

    Starting an academic career in research, by entering a PhD program

    More information at http://postgrau.lsi.upc.edu/?cmd=msc_intro

    PhD Programs

    PhD Program in Information Technology

    This PhD program provides scientific training for engineers and graduates within a specialized profile of research in information

    technologies: artificial intelligence, electronic commerce, communications, computer vision, automatic control, expert supervision

    and industrial robotics. This research corresponds to work carried out by research groups from the Department of Electronics,

    Computer Science and Automatic Control at the University of Girona.

    43Research Report 2006

  • 7/31/2019 Research Report 2006

    44/50

    More information at http://eia.udg.es/doctorat

    PhD in Software

    In general, the Software program addresses research areas related to the analysis, design and use of software. This means,in particular, research areas such as: computer-aided geometric design, virtual reality and synthesis of realistic images, information

    systems and databases, software engineering, programming methodology, theoretical computer science, analysis and design

    of algorithms and bioinformatics.

    In order to focus scientific and methodological skills on specific topics, the programme has a structure based around four core

    areas (corresponding to the four research sections of the Department): Algorithms and Complexity, Logic and Programming,

    Computer Graphics, and Information Systems. Each consists of a set of courses and seminars and offers supervised research

    projects within the corresponding research area.

    More information at http://ima.udg.es/Doctorat

    PhD Thesis

    Jaume Rigau, Information-Theoretic Refinement Criteria for Image Synthesis. Supervi-

    sors: Miquel Feixas and Mateu Sbert. December 2006.

    Luis Eduardo Mujica, A hybrid approach of Knowledge-Based Reasoning to Structural

    Assessment. Supervisor: Josep Veh. June 2006.

    Radu Oighidam, Catadioptric Stereo based on Structured Light Projection. Supervisor:

    Joaquim Salvi and El Mustapha Mouaddib. July 2006.

    Joan Jos Mora Flrez, Localizacin de faltas en sistemas de distribucin de energa

    elctrica usando mtodos basados en el modelo y mtodos basados en el conocimiento.

    Supervisors: Joan Colomer, Joaquim Melndez and Gilberto Carrillo. December 2006.

    Anna Urra, Multi-Layer Survivability: Routing Schemes for GMPLS-based Networks. Su-

    pervisors: Josep L. Marzo and Eusebi Calle. December 2006.

    Pau Herrero, Quantified Real Constraint Solving Using Modal Intervals With Applications

    to Control. Supervisor: Josep Veh and Luc Jaulin. December 2006.

    Diploma of Advanced Studies

    Fernando Solano, . Supervisor: . Date: July 18, 2006.

    Ronadl Uriel Ruiz Ordoez, . Supervisor: Josep Lluis de la Rosa. Date:July 18, 2006.

    Esteve del Acebo Pea, . Supervisor: . Date: October 16, 2006.

    Armagan Elibol, . Supervisor: . Date: October 16, 2006.

    Esteban Reinaldo Gelso, . Supervisor: . Date: October 16, 2006.

    Salvador Ibarra Martnez, . Supervisor: Josep Lluis de la Rosa. Date: October 16, 2006.

    Eduard Montaner, . Supervisor: Josep Lluis de la Rosa. Date: October 16, 2006.

    44 Research Report 2006

  • 7/31/2019 Research Report 2006

    45/50

    7. Competition7. Competition

    SAUC-E Team

    SAUC-E is the acronym for Student Autonomous Underwater Challenge - Europe.

    SAUC-E is a competition for European students to foment the investigation and development in underwater tech-

    nology.

    At the moment, there are 8 teams already inscribed in the competition:

    University of Bath

    International University Bremen

    University of Girona

    University of Glasgow

    Heriot-Watt University

    University of Leicester

    University of Nice Sophia-Antipolis

    Southampton University

    As the first event of its kind was held in the UK 2006, the Student Autonomous Underwater Challenge Europe

    (SAUC-E) proved a huge success. Seven university teams competed at the state-of-the-art underwater stage at

    Pinewood Studios (www.pinewoodshepperton.com) with vehicles they had designed and built themselves.

    For more information visit the official homepage: http://www.dstl.gov.uk/news_events/competitions/sauce

    45Research Report 2005

  • 7/31/2019 Research Report 2006

    46/50

    X. PublicationsX. Publications

    Books or Book Chapters

    Lpez, B., Melndez, J., Radeva, P. and Vitri, J. Artificial Intelligence Research and De-velpment, IOS Press (Serie:Frontiers in Artificial Intelligence and Applications), 2005,ISBN: 1-58603-560-6

    Gonzlez, G., Lpez, B., Angulo, C. and de la Rosa J.LL. Acquiring Unobtrusive Rele-vance Feedback through Eye-Tracking in Ambient Recommender Systems. Publishedin the book Artificial Intelligence Research and Development, 2005, ISBN: 1-58603-560-6

    B. Lpez, J.Melndez. Inteligencia Artificial en los planes de estudio de informtica e in-dustriales. Una perspectiva desde el rea de ISA. Published in the book DocenWeb:Red Temtica de Docencia en Control mediante Web. 2005, ISBN: 84-689-3386-4

    Vinyoles M., Melndez J., Herraiz S., Snchez J., Castro M. Location of phase to groundand phase to phase faults in a power distribution system. Published in the book 9 Con-greso Hispano-Luso de Ingeniera Elctrica (9CHLIE), 2005, ISBN: 84-609-5231-2

    Herraiz S., Melndez J., Colomer J., Vinyoles M., Snchez J., Castro M. Power qualitymonitoring in distribution systems. Published in the book 9 Congreso Hispano-Luso deIngeniera Elctrica (9CHLIE), 2005, ISBN: 84-609-5231-2

    Vinyoles M., Melndez J., Herraiz S., Snchez J., Castro M. Electric fault location me-thods implemented on an electric distribution network. Published in International Confe-rence on Renewable Energies and Power Quality (ICREPQ'05), 2005, ISBN:84-609-3234-6

    Ruiz M.L., Colomer J., Melendez. Combinaron of Statistical Process Control (SPC) me-thods and classification strategies for situation assessment of batch processes. Publishedin the book III Taller de Mineria e Datos y Aprendizaje (TAMIDA2005) dentro del CEDI05(congreso Espaol de Informatica), 2005, ISBN: 84-9732-449-8

    Lpez, B. , Urra, P., and Abell, X. Multi-agent resource allocation for road passengertransportation. Published in the book IV Workshop Internacional Sobre AplicacionesPrcticas de Agentes y Sistemas Multiagente (IWPAAMS), 2005, ISBN: 84-9773-222-7

    Lpez, B. , Melndez, J., Surez, S. Ontology for Integrating Heterogeneous Tools for Su-pervision, Fault Detection and Diagnosis. Published in the book Second InternacionalConference on Informatics in Control, Automation and Robotics (ICINCO), 2005, ISBN:972-8865-29-5

    B. Lpez, M. Montaner, J.L. de la Rosa. Utilizacin de un Simulador de Ftbol comoPlataforma de Prcticas de Inteligencia Artificial. Published in the book SImposio Nacional

    de Docencia en la Informtica (SINDI), 2005, ISBN: 84-9732-443-9Teran, MF, Colomer, J, Melndez J, Colprim J. Supervised Fuzzy Control of DissolvedOxygen in a SBR Pilot Plant. Published in the book Artificial Intelligence Research andDevelopment. Frontiers in in Artificial intelligence and Applications, 2005, ISBN: 1-58603-560-6

    Gonzlez, G., Angulo, C, Lpez, B. and de la Rosa, J. LL. Smart User Models for AmbientRecommender Systems. Published in the book Proceedings of the Workshop on Am-bient Intelligence and (Everyday) Life (Amilife), 2005, ISBN: 84-609-6632-1

    Josep Llus de la Rosa, Albert Figueras, Christian Quintero, Josep Antoni Ramon, and Sal-vador Ibarra. Computer Methods and Systems, 2005, ISBN: 83-916420-3-8

    G. Gonzlez, C. Angulo, B. Lpez, J.L. de la Rosa. Ambient Intelligence and (Everyday)Life, 2005, ISBN: ISBN 84- 609-6632-1

    Christian G. Quintero M., Josep Ll. de la Rosa and Josep Veh. Studies about the AtomicCapabilities Concept for Linear Control Systems in Physical Multi-Agent Environments.Published in the book Proceedings 2005 IEEE International Conference on Computatio-nal Intelligence in Robotics and Automation, 2005, ISBN: 0-7803-9355-4

    Christian G. Quintero M., Salvador Ibarra M., Josep Ll. de la Rosa and Josep Veh. Ex-ploring the Physical Agents' Behaviours to Improve Collective Decisions in Multi-agent En-vironments. Published in the book Actas VI Workshop en Agentes Fsicos. CongresoEspaol de Informtica 2005, ISBN: 84-9732-450-1

    Christian G. Quintero M., Josep Ll. de la Rosa and Josep Veh. Self-Knowledge basedon the Atomic Capabiliites Concept. A Perspective to Achieve Sure Commitments amomgPhysical Agents. Published in the book Proceedings 2005 International Conference onInformatics in Control, Automation and Robotics, 2005, ISBN: 972-8865-30-9

    S.Aciar, J. Guzman-Obando, J.LL. de la Rosa and J. Lpez Herrera. Estableciendo unamedida de Trust para seleccionar fuentes de informacin relevantes para recomendar.Published in the book Actas Cuarto Congreso de Cmputo AGeComp-UAEM, 2005,ISBN: 968-878-250-0

    S. Aciar , J. Lpez Herrera y J.LL. De la Rosa. SMA para la bsqueda inteligente de in-formacin para recomendar. Published in the book Actas del Simposio de InteligenciaComputacional, SICO2005 (IEEE Computational Intelligence Society, (SC), 2005, ISBN:84-9732-444-7

    Josep Veh, Jorge Bonda. El anlisis de intervalos en Espaa: desarrollos, herramientasy aplicaciones 2005, ISBN: 84-934349-1-4

    Veh Josep, Ferrer Ins, Sainz Miguel A. Aplicaciones del anlisis intervalar al control ro-busto. Published in the book El anlisis de intervalos en Espaa: desarrollos, herramien-tas y aplicaciones, 2005, ISBN: 84-934349-1-4

    Joaquim Armengol, Pau Herrero, Miguel ngel Sainz. Deteccin de fallos en procesosreales (Proyecto CHEM) . Published in the book El anlisis de intervalos en Espaa:desarrollos, herramientas y aplicaciones, 2005, ISBN: 84-934349-1-4

    Gabriela Caldern, Pedro J. Abad, Arantza Aldea, Joaquim Armengol. Diagnstico de fa-llos. Published in the book El anlisis de intervalos en Espaa: desarrollos, herramientasy aplicaciones, 2005, ISBN: 84-934349-1-4

    Joaquim Armengol, Vicen Puig, Joseba Quevedo, Teresa Escobet. Simulacin interva-lar y su aplicacin a la deteccin robusta de fallos en sistemas dinmicos con incertidum-bre paramtrica. Published in the book El anlisis de intervalos en Espaa: desarrollos,herramientas y aplicaciones, 2005, ISBN: 84-934349-1-4

    Luis Eduardo Mujica, Josep Veh and Jos Rodellar. A hybrid system combining SelfOrganizing Maps with Case Based Reasoning in Structural Assessment in Artificial Inte-lligence Research and Development. Published in the book Frontiers in in Artificial inte-lligence and Applications, 2005, ISBN: I-58603-560-6

    Christian G. Quintero M., Josep Ll. de la Rosa, Josep Veh. Control Inteligente en AgentesFsicos. Una Propuesta basada en Capacidades Atmicas. Simposio Espaol sobreControl Inteligente, 2005, ISBN: 84-689-2599-3Luis Eduardo Mujica, Josep Veh and Jos Rodellar. Deteccin de impactos medianteRazonamiento Basado en Conocimiento: Aplicacin a una seccin de ala de avin. Pu-blished in the book Actas de SAICA 2005 Seminario de Aplicaciones Industriales de Con-trol Avanzado, 2005, ISBN: 84-362-5179-2

    Christian G. Quintero M., Salvador Ibarra M., Josep Ll. de la Rosa and Josep Veh. Ex-ploring the Physical Agents Behaviours to Improve Collective Decisions in Multi-agent En-vironments. Published in the book Congreso Espaol de Informtica, CEDI 2005, ISBN:84-9732-450-1

    Josep Ll. de la Rosa, ALbert Figueras, Christian G. Quintero M., Josep A. Ramn and Sal-vador Ibarra. Outline of Modifications Systems Computer Methods and Systems. 2005,ISBN: 83-916420-3-8

    Christian G. Quintero M., Josep Ll. de la Rosa and Josep Veh. Self-Knowledge basedon the Atomic Capabilities Concept. A Perspect