Sitemap Sitemap | Search Search | Lock Intranet

Control Engineering and Intelligent Systems

Research projects

List of current and past research projects within the group:

Active projects

PEPPER – Patient Empowerment through Predictive PERsonalised decision support.
Principal investigator: Beatriz López.
Partners: Oxford Brookes University, Imperial College of Science, Technology and Medicine, Universitat de Girona, Institut d’Investigació Biomèdica de Girona, Romsoft SRL and Cellnovo Limited..
Dates: 01/Feb/2016 - 31/Jan/2019.
Sponsor: European Commission.
Reference: PHC-28-2015, Grant Agreement 689810.
Description: PEPPER is a personalised decision support system for chronic disease management that will make predictions based on real-time data in order to empower individuals to participate in the self-management of their disease.

The design will involve users at every stage to ensure that the system meets patient needs and raises clinical outcomes by preventing adverse episodes and improving lifestyle, monitoring and quality of life. Research will be conducted into the development of an innovative adaptive decision support system based on case-based reasoning combined with predictive computer modelling. The tool will offer bespoke advice for self- management by integrating personal health systems with broad and various sources of physiological, lifestyle, environmental and social data. The research will also examine the extent to which human behavioural factors and usability issues have previously hindered the wider adoption of personal guidance systems for chronic disease self-management. It will be developed and validated initially for people with diabetes on basal-bolus insulin therapy, but the underlying approach can be adapted to other chronic diseases.

There will be a strong emphasis on safety, with glucose predictions, dose advice, alarms, limits and uncertainties communicated clearly to raise individual awareness of the risk of adverse events such as hypoglycaemia or hyperglycaemia. The outputs of this research will be validated in an ambulatory setting and a key aspect will be innovation management. All components will adhere to medical device standards in order to meet regulatory requirements and ensure interoperability, both with existing personal health systems and commercial products. The resulting architecture will improve interactions with healthcare professionals and provide a generic framework for providing adaptive mobile decision support, with innovation capacity to be thereby increasing the impact of the project.

HIT2GAP: Highly Innovative building control Tools Tackling the energy performance GAP.
Principal investigator: Joaquim Meléndez.
Dates: 01/Sep/2015 - 31/Aug/2019.
Sponsor: CE.
Reference: H2020 Grant Agreement No.: 680708 .
Description: Measurement campaigns have shown major discrepancies in buildings energy performance between planned energy demand and real energy consumption, while nowadays most of the newly constructed offices buildings are equipped with BMS systems, integrating a more or less extended measurement layer providing large amounts of data. Their integration in the building management sector offers an improvement capability of 22 % as some studies demonstrate.
The HIT2GAP project will develop a new generation of building monitoring and control tools based on advanced data treatment techniques allowing new approaches to assess building energy performance data, getting a better understanding of building’s behaviour and hence a better
performance. From a strong research layer on data, HIT2GAP will build on existing measurement and control tools that will be embedded into a new software platform for performance optimization.

The HIT2GAP solution will be applied as a novel intelligent layer offering new capability of the existing BMS systems and offering the management stakeholders opportunities for services with a novel added value. Applying the solutions to groups of buildings will also allow to test energy demand
vs. local production management modules. This will be tested in various pilot sites across Europe. HIT2GAP work will be realized with a permanent concern about market exploitation of the solutions developed

ACCUS: Adaptive Cooperative Control in Urban (sub) Systems.
Principal investigator: Lluïsa MArsal and Joaquim Meléndez.
Partners: Inabensa, TNO, Hi Iberia, GUT, Thales, UPM, etc.
Dates: 01/Jun/2013 - 31/May/2016.
Sponsor: Ministry of Industry, Energy and Tourism and JTI ARTEMIS Joint Undertaking of the European Commision.
Reference: ART-010000-2013-2 -333020-1.
The ACCUS (Adaptive Cooperative Control in Urban (sub)Sytems) is an ambitious Project meant to achieve a better sustainable management and to engage synergies in order to improve citizens’ quality of life. It is motivated by the societal challenge “Smart buildings and communities in the future” and the supporting scenarios “Mobility for everyone”,”Energy control in the urban and rural environment” and “Security” as defined in the ARTEMISIA Strategic Research Agenda (SRA).

The ACCUS project is focused on the integration of urban subsystems such an intelligent transport systems, city light systems and energy management systems. Rather than developing solutions for specific applications and scenarios, the ACCUS project will develop theoretical framework (e.g. semantic interoperability concepts) and practical framework (e.g. methodology, reference architecture) to design new applications within converged scenarios.

MoSHCA - My Mobile and Smart Health Care Assistant.
Principal investigator: Beatriz López.
Partners: Planet Media, Answare Tech, Institut d'Investigació Biomèdica de Girona Dr. Josep Trueta, Actimage, Evalan, Radboud University,Instituto de Telecomunicaçõe,etc.
Dates: 01/Oct/2012 - 31/Dec/2015.
Sponsor: MINECO.
Reference: EUREKA ITEA 2 nº 11027 - IPT-2012-0943-300000.
Description: The main objective of MoSHCA is to develop intelligent healthcare system solutions for patients with chronic diseases(diabetes, epilepsy, pregnancy related disorders, etc ) and general health monitoring (baby monitoring and general health) through medical sensors and smartphones. The eXiT group is in charge of novel intelligent algorithms, decision making models and alerts in baby monitoring domain.

Past projects

BENEFIT: Better Effectiveness aNd EFficiency by measuring and modelling of Interventional Therapy .
Principal investigator: Joaquim Melendez and Imma Boada.
Partners: Sorin Group, OSM, MERKUM, Hospital Clinic de Barcelona.
Dates: 01/Nov/2014 - 31/Dec/2016.
Sponsor: MINETUR- Acción Estratégica de Economía y Sociedad Digital.
Reference: TSI-100103-2014-89.
Description: BENEFIT will improve patient stratification and patient treatment by quantification and integr ation of
pre- and intra-interventional images/ data into personalized models for workflow and CDSS

MESC: Platform for Monitoring and assessing the Efficiency of distribution systems in Smart Cities.
Principal investigator: Carles Pous and Joan Colomer.
Partners: University of Girona (UdG) University Polytechnic of Madrid (UPM).
Dates: 01/Jan/2014 - 31/Dec/2016.
Reference: DPI2013-47450-C2-1-R.
Description: Coordinated project, participated by two research groups: eXIT(Control Engineering and Intelligent Systems) from the Universitat de Girona and SYST (System & Software Technology) from the Universidad Politecnica de Madrid (UPM), that aims contributing to improve efficiency of distribution systems in smart cities by means of the application of advanced monitoring.

The project envisions a platform to support monitoring procedures for distribution systems in Smart Cities. This requires a software engineering vision to specify a reference architecture that supports variability for different domains (energy, water, city mobility, etc.), vendors, etc. Under this perspective the consortium combines expertise of UdG on data driven monitoringmethods (modeling, fault detection and diagnosis and benchmarking) with excellence of UPM on software engineering.

SEAS: Smart Energy Aware Systems.
Principal investigator: Sergio Herraiz and Joaquim Meléndez.
Partners: Answare Tech, Planet Media, Abada, Igeniería y Control. .
Dates: 01/Sep/2013 - 31/Dec/2014.
Sponsor: MINETUR.
Reference: TSI-100501-2013-47.
Description: The project objective includes the introduction of new digital services enabling proactive adaptation of consumption behaviour to the changes in usage, energy costs, availability of energy sources and weather/climate/season, while making the stakeholders (e.g. utilities and consumers) aware of the changes and their impact.

SEPS - Sistema experto de probabilidad y severidad de incidentes de red.
Principal investigator: Sergio Herraiz.
Partners: Union Fenosa Distribución, Telvent Energía, AIA, SCIEN Analitics, Univ. Carlos III de Madrid, Universidad de Zaragoza.
Dates: 01/Jan/2013 - 31/Dec/2014.
Sponsor: MINECO.
Reference: IPT-2012-0580-120000.
Description: The main objective of SEPS is to develop an expert system to generate occurrence probability indexes regarding events, incidents and constraints in the power distribution grid. Weather forecast and demand/generation forecast will be used in order to obtain these indexes. Different algorithms will be developed to identify the severity and social impact of these events, taking into account both intrinsic (power, affected clients, recovery time, etc) and extrinsic(social sensitivity, relevant events, etc) aspects.

Principal investigator: Beatriz López.
Partners: UdG.
Dates: 01/Jan/2012 - 31/Dec/2012.
Sponsor: Ministerio de Ciencia e Innovación.
Reference: DPI2011-24929.
Description: -

DICOMA - DIsaster COntrol MAnagement.
Principal investigator: Joaquim Meléndez.
Partners: Siemens. INDRA Software Labs.,GEO Concept, Telecom Bretagne, Thales, Forthnet, Arttic, Space Applications, Answare Tech, CreativIT, Goodmill Systems, Mattersoft, Mobisoft, SAVOX, Sec-Control Group, Specil Code, Sunit: AthenaGS3, RUNCOM, C2Tech, Mantis, UASD, University of Paderborn, Uhasselt, VUB, Universidad Autónoma de Barcelona, Universitat de Girona, UPC, Universidad de Sevilla, ATB, FDB, FDDO, THW, Tecnológico Fundación Deusto, CNRS, Loria-Inria, FMI, OUAS, VTT..
Dates: 15/Dec/2011 - 30/May/2013.
Sponsor: Ministerio de Industria Turismo y Comercio.
Reference: Eureka Itea 2 no10031- TSI-020400-2011-55.
Description: The goal of the DiCoMa project is to ensure effective management of large disasters and complex emergencies by providing a set of tools that aim to improve the effectiveness of decision makers in dealing with disasters by better training and in situ support in the field. This toolset will include Data Abstraction tools, Simulation and Modelling Tools and Decision Support and Training tools

ENERFICIENCY - User Led Energy Efficiency Management.
Principal investigator: Joaquim Meléndez.
Partners: INDRA Software Labs, ZeroEmissions, EADS, ABB, CreativIT, E&D, KEMA Energy, HPC-SA, CEA, Aeltari International, De Clercq Solutions, Delta Engineering, Freemind, 3E, IT Synergy, NRG-Solutions, Silminds, Raiden NRG, MHG, Universidad Carlos III Madrid, Universitat de Girona, Universitat Jaume I , Tallin Technical University, Sirris, VTT.
Dates: 01/Dec/2011 - 28/Feb/2014.
Sponsor: Ministerio de Industria Turismo y Comercio.
Reference: EUREKA ITEA 2 no 10009 - TSI-020400-2011-56 .
Description: The ENERFICIENCY project is an initiative to design, implement and test an open software platform for energy efficiency monitoring and management from the customer (consumers/prosumers) side capable to interact with the power network and to provide services for an efficient use of energy.
The project aims to build a software platform capable to integrate many and heterogeneous data sources and to provide appropriate services to convert these data into valuable information for energy efficiency management, audit and benchmarking. Development efforts will be distributed in four main topics: software architecture, communications, data acquisition/management and computational intelligence.

IMPONET - Intelligent Monitoring of POwer NETworks.
Principal investigator: Joaquim Meléndez.
Partners: Indra Software Labs(Spain) Indra Sistemas (Spain) Mnemo (Instituto Tecnológico de León)(Spain) Answare Tech (Spain) Prodevelop (Spain) DeustoTech (Spain) Tecnalia (European Software Institute)(Spain) KEMA (Spain) Unión Fenosa Distribución (Spain) Universitat de Girona (Spain) UPM (Spain) Vrije Universiteit Brussel (Belgium) Capvidia (Belgium) ETFOS - Faculty of Electrical Engineer (Croatia) HEP Operator Prijenosnog Sustava (Croatia) Cairo Orange Labs (Egypt) EOIP (Egypt) Saba Electric (Egypt) Nile University (Egypt) Kapion (Slovenia) University of Ljubljana (Slovenia) WOOAM Inc.(South Korea) LNL Technology (Turkey) INNOVA (Turkey) .
Dates: 01/Dec/2010 - 28/Feb/2013.
Sponsor: Ministerio de Industria Turismo y Comercio.
Reference: Eureka ITEA2 no 09030 - TSI-020400-2010-103.
Description: The main goal of the IMPONET project is to focus on the strategic and technical requirements being demanded by the electrical markets worldwide regarding: smart grids, smart metering, peak demand management, demand response and real time monitoring. In order to fulfil this main goal, the scope of the IMPONET project is to model, design and implement a comprehensive, flexible and configurable information system, in order to meet the most complex and advanced requirements in Energy Management.

MEDIATE - Patient Friendly Medical Intervention.
Principal investigator: Joaquim Meléndez.
Partners: The Netherlands: Philips Healthcare Philips Research Nucletron Prodrive Technolution Demcon Amsterdam Medical Centre (AMC) Erasmus MC Leiden UMC (LUMC) UMC Utrecht (UMC) Belgium: Barco IBBT France: Cedrat Digisens EndoControl Haption Philips Medisys SurgiQual Institute Therenva Institut Telecom CEA List LTSI Spain: Alma IT Systems Atos Origin IAC3 VICOMTech Universitat de Girona (UdG) .
Dates: 01/Sep/2010 - 31/Dec/2013.
Sponsor: Ministerio de Industria Turismo y Comercio.
Reference: Eureka ITEA 2 no 09039 - TSI-020400-2010-84.
Description: The objective of the Mediate project is to increase productivity and effectiveness in healthcare and reduce patient risk and discomfort by supporting healthcare professionals in the transition from invasive, open surgery to minimally invasive, image guided intervention and treatment (IGIT). By empowering the healthcare professional through more advanced technologies during the whole treatment cycle, IGIT helps them to obtain a better clinical outcome of the treatment, predictable procedure times, fewer complications, better service to the patient and lower morbidity & mortality rates.

The final outcome of the Mediate project are clinical demonstrators for different disease areas (cardiology, oncology and orthopaedics) that will incorporate advanced technologies in image generation & analysis, procedural navigation & guidance, decision support systems and workflow management.

eXiT*DM - Data Mining Platform.
Principal investigator: Joaquim Meléndez.
Partners: -.
Dates: 01/Apr/2010 - 31/Mar/2011.
Sponsor: UdG.
Reference: singular_udg.
Description: This project intends to develop a platform that integrates complex data mining tools and also a data generation engine that will allow better experimentation. New algorithms will be attached to the platform as plugins so that it will have a modular structure. Once the application is done, we will post it on our website ( as open source so that anyone will be able to work with it.

i-PQM - Monitorización inteligente de la calidad de la energía eléctrica .
Principal investigator: Joaquim Meléndez.
Dates: 01/Jan/2010 - 31/Dec/2012.
Reference: DPI2009-07891.
Description: This project has been funded by the Ministry of Science and Innovation of the Spanish Government (project ref. DPI2009-07891.) and it is conceived as the continuity of a previous one (DPI2006-09370, “Diagnóstico de redes eléctricas de distribución basado en casos y modelos”). The aim is to build an intelligent power quality monitoring framework to exploit temporal and spatial correlation of disturbances gathered in multiple monitoring points around the power network to assist maintenance polices. Partial objectives of the project cover the application of data mining strategies to discover sequence patterns useful for the improvement of fault location methods and prognosis of incipient faults in the distribution network; and also the exploitation of spatial correlation of data registered by multiple power quality registers scattered around the power network.

ENERGOS - Tecnologías para la gestión automatizada e inteligente de las redes de distribución energética del futuro.
Principal investigator: Unión Fenosa - UdG: J. Meléndez.
Partners: Unión Fenosa, Indra and about 30 more..
Dates: 16/Sep/2009 - 31/Dec/2012.
Sponsor: CDTI-CENIT.
Reference: CEN20091048.
Description: This project aims to do research on the basic elements that form a smart grid able to manage in real time all multidirectional energy flows that will have the power grid of the future. This grid will be provided with many renewable energy sources at different levels and also will give the customers higher control over their energy management. It will be a more efficient grid and will also contemplate the electric car of the future which will consume but also supply power to the grid.

Infraestructura Monitorización Calidad Onda Campus Montilivi.
Principal investigator: Sergio Herraiz.
Dates: 01/May/2009 - 01/Jul/2009.
Reference: PEIR-DGR2007.
Description: Adquisición e instalación de 13 equipos de medida de calidad de onda para la monitorización del Campus de Montilivi de la Universitat de Girona.

RDS granular de una etapa para EBN: diseño y operación con un sistema de control avanzado.
Principal investigator: Joan Colomer.
Dates: 01/Jan/2009 - 01/Dec/2011.
Reference: CTQ2008-06865-C02-02.
Description: The GRASTAC project aims to advance in the biological nutrient removal from wastewater by the application of the emerging granular technology applied to SBR. In this respect this project intends to identify the basic knowledge about the granulation process (microbial activity of granular biomass for nutrient removal, population dynamics distribution and physical/chemical composition evolution of biological granules during the granule development and operation with variable composition wastewater. Apart from the basic knowledge, this project also intends to develop a new tubular granular SBR for nutrient removal with synthetic wastewater for a better process stability and performance. In this respect a lab scale pilot plant (30 litres) will be designed, constructed, started-up and operated for biological nutrient removal process. Afterwards, the granular technology will be applied to an available 1000 litres industrial pilot plant for technology validation in a full scale WWTP. On the other hand, in order to ensure process stability and performance in nutrient removal, the previously developed SICOTIN software will be adapted to the granular technology by including new data acquisition probes, adapting the advanced control modules of Case Based Reasoning (CBR) and Multivariable Statistical Process Control (MSPC) module to the granular technology. Also the advanced control module will implement a new Software Sensor module for the on-line identification of the key chemical compositions of ammonium and phosphate in order to achieve a more robust process control able to deal with the variable influent compositions while keeping the performance in nutrient removal from wastewater.

SUROS - Subastas robustas mediante la incorporación de técnicas de satisfactibilidad módulo teorías.
Principal investigator: Dídac Busquets.
Dates: 01/Jan/2009 - 01/Dec/2011.
Sponsor: none.
Reference: TIN2008-04547.
Description: Robust auctions through the incorporation of satisfiability module theories techniques Auctions are mechanisms for distributing goods among a set of agents.

Usually, it is assumed that the result of the auction is a binding contract that has to be fulfilled by the agents. However, this assumption is not valid when the environment in which the auctions are used is a real one, since unexpected events could occur, for which the solutions may not be prepared. Thus, in this project we will focus our research in an innovative field: robust auctions. To this end, we must clearly define what is meant by robustness in the auction domain, as well as develop methods for providing the auctions with a robust behavior. More concretely, we will develop robust auction mechanisms, which should incentivize agents to fulfill the contracts, and also algorithms for generating robust solutions, which should be easily repairable in the case of unexpected events. Moreover, we plan to use logic based techniques, in particular those of satisfiability modulo theories (SMT) to tackle combinatorial problems, such as the auction winner determination problem. Such techniques are highly expressive, and recently there have been huge advances in the efficiency of the solvers. Therefore, in the project we will design logic based methods for solving auctions and other similar combinatorial problems. Moreover, we will also generalize the concept of robustness within the SMT domain.

AIMES - Advanced infrastructure for medical equipment management and services.
Principal investigator: Beatriz López.
Partners: Dräger Medical AG & Co. KG, Enverdis GmbH, FEIG ELECTRONIC GmbH, ifak e.V. Magdeburg, INDRA Sistemas, RGB Medical Devices S.A., SBSK GmbH & Co. KG Siemens AG, SUINSA Medical Systems S.A., TELEFONICA I+D, Universitätsklinikum Magdeburg, University of Girona..
Dates: 01/May/2008 - 01/Nov/2010.
Reference: ITEA2 ~ 07017 / TSI-020400-2008-47.
Description: AIMES is aimed at developing a service infrastructure for the overall technical hospital facilities. New tools for management, distributed condition monitoring and diagnosis of fix and mobile medical equipment will be integrated into the appropriate communication infrastructure.

Principal investigator: ---.
Dates: 01/Apr/2008 - 01/Dec/2008.
Reference: -.
Description: The purpose of this collaboration has been oriented to develop a new diagnosis methodology for the ARIANE engine based on multivariate statistical process control principles. Main improvements are in the selection of monitored variables and number of principal components to build the mode and the selection of statistical thresholds in the contribution analysis for diagnosis.

Diagnóstico de redes eléctricas de distribución basado en casos y modelos (DCM-RED).
Principal investigator: Joaquim Meléndez.
Dates: 01/Oct/2006 - 31/Dec/2009.
Reference: DPI2006-09370.
Description: This project has been funded by the Ministry of Education and Science of the Spanish Government (project ref. DPI2006-09370). The aim is to improve power quality monitoring methods to assist the diagnostic of disturbances gathered by power quality monitors with emphasis in those produced by fault occurring in the distribution network. A combination of data driven methods and models based approaches is proposed to assist fault location in Distribution Networks. Case based reasoning (CBR) strategy is proposed to diagnose electric faults based on the association between events gathered by power quality monitors and incidents reported in the network control center. Network models are used in the fault location step to estimate the distance to the fault. The project has been developed in collaboration with ENDESA DISTRIBUCIÓN.

Aplicación de ayuda a la monitorización de perturbaciones y localización automática de faltas en redes de distribución eléctrica.
Principal investigator: Joaquim Meléndez.
Partners: UdG, ENDESA.
Dates: 01/Jun/2006 - 30/May/2007.
Sponsor: MEC.
Reference: PTR95-1020.OP.
Description: Herramienta para la ayuda a la localización de fallos en redes de media tensión A partir de medidas proporcionadas por equipos de calidad de onda. Incluye la caracterización y clasificación de perturbaciones y le estimación del punto de falta para las ocurridas en media tensión.

Aplicación de ayuda a la monitorización de perturbaciones y localización automática de faltas en redes de distribución eléctrica.
Principal investigator: ---.
Dates: 01/Jun/2006 - 30/May/2007.
Reference: PTR95-1020.OP.
Description: Herramienta para la ayuda a la localización de fallos en redes de media tensión A partir de medidas proporcionadas por equipos de calidad de onda. Incluye la caracterización y clasificación de perturbaciones y le estimación del punto de falta para las ocurridas en media tensión.

Principal investigator: Belarmino Pulido ( U. de Valladolid).
Partners: UValladolid, USevilla, UPC, UdG.
Dates: 01/Jan/2006 - 01/Jan/2006.
Sponsor: MEC.
Reference: DPI2005-23815-E.
Description: ---

KD-Ubiq - A blueprint for ubiquitous knowledge discovery systems.
Principal investigator: Michal May.
Dates: 01/Dec/2005 - 30/May/2008.
Reference: FP6-2002-IST-C.
Description: KDubiq (Knowledge Discovery in Ubiquitous Environments)
is a CA (coordination action). This means we will rely on the active network members from KDNet and other NOEs and concentrate on integration of embedded systems community and joint-industry activities. The project will support international workshops and conferences, working groups and technical meetings, KDubiq Portal applications, etc. The last and most comprehensive result of the project will be a book “Blueprint on Ubiquitous Knowledge Discovery” about the state-of-the-art in data mining applications and technology for ubiquitous environments. At the end of the project period, each Working Group (see below) will deliver a comprehensive chapter to this book.

Aplicabilidad de algoritmos para la localización de fallos en redes de distribución eléctrica.
Principal investigator: ---.
Dates: 01/Feb/2005 - 01/Feb/2006.
Reference: ---.
Description: Estudiar la viabilitat i precisió dels algoritmes descrits en la literatura per localitzar falles reals en xarxes de distribució.

Principal investigator: ---.
Dates: 01/Jan/2005 - 01/Jan/2008.
Reference: 225997-CP-1-2005-1-FR-ERASMUS-TNPP.
Description: ---

Red Temática Española de Agentes y Sistemas Multi-Agente.
Principal investigator: A. Moreno.
Partners: 15 research centers.
Dates: 01/Jan/2005 - 01/Jan/2005.
Sponsor: MEC.
Reference: ---.
Description: ---

Desarrollo y planificación de técnicas de resolución de problemas de scheduling en entornos multiage.
Principal investigator: Beatriz López.
Partners: UdG, UPV.
Dates: 13/Dec/2004 - 12/Dec/2007.
Sponsor: MEC.
Reference: TIN2004-06354-C02-02.
Description: En este proyecto se pretende desarrollar técnicas de Computación Distribuida y Sistemas Multiagente para la resolución de Problemas de Satisfacción de Restricciones (CSP), particularmente orientados hacia procesos de planificación y scheduling (asignación de recursos compartidos). Por ello, se trata de un proyecto coordinado en el que confluyen dos grupos de investigación con experiencia complementaria en: (i) Técnicas CSP,CSP-distribuidos, planificación y scheduling y (ii) entornos distribuidos y abiertos, sistemas multi-agente, y CSP-multiagente.
Así mismo, se tratará la integración de los procesos de Planificación y Scheduling bajo una consideración común de
técnicas CSP. El desarrollo de técnicas CSP-Multiagente, así como la integración de los procesos de planificación y
scheduling, son objetivos científicamente relevantes, ambiciosos y aplicables, permitiendo tratar una amplia tipología de problemas en los que las técnicas tradicionales CSP, así como los esquemas tradicionales de Planificación & Scheduling, son difícilmente aplicables y/o ineficientes.
Adicionalmente, el proyecto de investigación pretende aplicar y evaluar las técnicas desarrolladas en escenarios de evaluación ampliamente contrastados en la comunidad científica y, sobre todo, con dos aplicaciones concretas y reales de transporte ferroviario y transporte interurbano, en las que se dispone de respectivos convenios de colaboración.
La capacidad científica de los equipos es contrastable por los resultados científico-técnicos obtenidos en anteriores
proyectos de colaboración, el último de ellos conjunto, así como por la participación en proyectos y redes europeas de
temática similar y las relaciones con otros grupos nacionales e internacionales.

Monitorització i Control Centralitzat d'una conca de sanejament.
Principal investigator: Andreu Catala (UPC) / UdG: Joan Colomer.
Partners: CETENASA, UPC, CNRS, UdG, UTPerpignan.
Dates: 28/Dec/2003 - 30/Sep/2004.
Sponsor: DURSI.
Reference: PCI-ITT2002-3.
Description: Monitorització d'un riu i anàlisi de d'influència dels aports efectuats des de EDARs.

Pronòstic consum d'aigües a Girona.
Principal investigator: Joaquim Melendez.
Partners: Aigues de Girona, UdG.
Dates: 01/Oct/2003 - 28/Feb/2004.
Sponsor: Aigües de Girona, Salt i Sarrià de Ter.
Reference: Conveni Aigues 2003.
Description: Pronòstic de consum d'aigua mitjançant Xarxes Neurals. Comparativa.

Managements of events and incidences registered in a substation. Fault location in MT.
Principal investigator: Joaquim Melendez.
Partners: Endesa Distribucion, UdG.
Dates: 23/Jul/2003 - 02/Jul/2004.
Reference: Contracte ENDESA 2003.
Description: Desenvolupament d'una aplicació per la conciliació d'events registrats en la subestació de Salt i les incidències del centre de control (extrets de la BDE). Implementació de algorismes de localització de falles en xarxes de MT radials.

Red Española de Minería de Datos y Aprendizaje Automático.
Principal investigator: Jose Riquelme.
Partners: 17 research groups from 15 Spanish universities.
Dates: 01/May/2003 - 01/May/2004.
Sponsor: MCYT.
Reference: CICYT TIC2002-11124-E.
Description: Red temática en minería de datos.
Libro de la red (provisional):

Computación mediante restricciones para robótica y gestión de recursos.
Principal investigator: Carme Torres (IRII).
Partners: CSIC (IRII, IIIA), UPC.
Dates: 01/Jan/2003 - 01/Jan/2004.
Sponsor: CICYT.
Reference: TAP-1999-1086-C03.
Description: ---

Red Nacional de Planificación, Scheduling y Razonamiento Tempora.
Principal investigator: Luis Castillo (Universidad de Granada).
Dates: 01/Jan/2003 - 01/Jan/2004.
Sponsor: MEC.
Reference: ---.
Description: ---

Desarrollo de un sistema de control y supervisión aplicado a un reactor secuencial por cargas (SBR) para la eliminación de materia orgánica, nitrógeno y fósforo.
Principal investigator: Joan Colomer.
Partners: UdG (Lequia /EXIT).
Dates: 01/Dec/2002 - 30/Nov/2005.
Sponsor: MCYT.
Reference: DPI2002-04579-C02-01.
Description: ---

Power Quality monitoring: Classification and analysis of sags in a 25kV substation.
Principal investigator: Joaquim Melendez.
Partners: ENDESA, UdG.
Dates: 15/Feb/2002 - 15/Feb/2003.
Reference: Contracte ENDESA 02-03.
Description: Anàlisis de formes d'ona de sots de tensió. Obtenció d'atributs temporals freqüencials i fasorials. Classificació automàtica de pertorbacions.

Supervisión experta de la calidad de servicio eléctrico.
Principal investigator: Joaquim Meléndez.
Partners: UdG, UIS (colombia).
Dates: 28/Dec/2001 - 27/Dec/2004.
Sponsor: MCYT.
Reference: DPI2001-2198.
Description: Evaluación de la Calidad de Onda a partir del análisis de la señal registrada. Minería de datos aplicada a bases de datos con perturbaciones, clusterización y clasificación de perturbaciones en base a atributos temporales.

Diseño y Planificación de Agentes Físicos Dinámicos.
Principal investigator: Josep Lluís de la Rosa.
Partners: UdG, URV, UPV.
Dates: 28/Dec/2001 - 27/Dec/2004.
Sponsor: MCYT.
Reference: DPI2001-2094-C03-01.
Description: Desarrollamos tecnología apta para sacar provecho de múltiples y heterogéneos agentes físicos y software. En el proyecto ejemplificamos la investigación sintetizando agentes físicos como robots móviles en su aplicación a tareas de interés social, como es el rescate de personas en catástrofes como terremotos, incendios, inundaciones, y guerras. El rescate es un tema de interés mundial, en el que hasta ahora se ha aplicado poca informática, y el componente de ejecución humano es esencial, en el que se desconoce las posibilidades de aplicación de las nuevas técnicas de robótica móvil e inteligencia artificial.
¿Existe la tecnología para coordinar robots en tareas de rescate? NO. Tampoco existen desarrollos concretos eficaces para sacar jugo de muchos robots en tareas cooperativas. La tecnología a desarrollar consiste en superar el conocimiento existente de las propiedades de los sistemas heterogéneos, tal como son los sistemas multirobot móviles cooperantes, donde cada robot es diferente por construcción, y donde cada robot desarrolla tareas específicas como la localización, extracción, logística y transporte de personas, provisiones y tareas de soporte al rescate. Es evidente que para dichas tareas se necesita usar múltiples robots; así se mejora el coste, robustez y la eficacia en salvar vidas.
La investigación va a superar el estado del arte en los tres aspectos básicos siguientes:
• ¿Cómo manejar la heterogeneidad de robots, donde cada robot ha sido diseñado en países distintos?
• ¿Cómo conseguir que los multirobots trabajen coordinadamente de forma planificada dentro de un entorno tan desestructurado como una ciudad en ruinas, teniendo en cuenta que son robots heterogéneos?
• ¿Qué elementos robóticos han de incorporar, entre otros la visión, el olor y la parte motora, para adaptarse mejor a las tareas cooperativas?
Respecto a la primera pregunta, se abren nuevos horizontes de investigación sobre la importancia de los ecosistemas de agentes físicos para el análisis de sistemas multirobots móviles cooperantes, donde se aprecia la necesidad de una correcta diversidad (heterogeneidad) del conjunto robótico. La investigación está muy incipiente en este tema. Para abordar la segunda pregunta es necesario el desarrollo de nuevas técnicas de Planificación en Inteligencia Artificial para la resolución de problemas reales en entornos dinámicos, no deterministas y multi-agente donde se comparten recursos. Adicionalmente, habrá que diseñar técnicas de replanificación para el proceso de monitorización de la ejecución de los planes. Respecto a la tercera pregunta se necesita estudiar la integración informática y perceptual de dichos robots con la finalidad última de obtener un modelo tridimensional que refleje el estado de una escena no estructurada. Este modelo se obtendrá de manera distribuida a partir de sensores montados sobre un equipo de robots móviles que han de desplegarse por la escena de trabajo de forma coordinada. Los sensores considerados son cámaras de color de bajo coste, sensores de distancia por ultrasonidos y cámaras 3D pasivas.
Dicha investigación se ejemplificará con múltiples robots de las universidades participantes, en entornos emulados de rescate. Los resultados serán de aplicación general y de alto impacto social.

Advanced Decision Support Systems for Chemical/Petrochemical.
Principal investigator: Josep Lluís de la Rosa / Joan Colomer.
Dates: 01/Apr/2001 - 31/Mar/2004.
Sponsor: European Union.
Reference: ---.
Description: The aim of the CHEM project is to develop and implement an advanced Decision Support System (DSS) for process monitoring, data and event analysis, and operation support in industrial processes. The system will be a synergistic integration of innovative software tools, which will improve the safety, product quality and operation reliability as well as reduce the economic losses due to faulty states, mainly in refining, chemical and petrochemical processes.
Exit group contribuion is focused on the devlopment of ToolBoxes fro Qualitative representation of trends (http://exit/qualtras.aspx)and classification of situations.

Monitorització Intel.ligent Preventiva d'Estacions Depuradores d'Aigua.
Principal investigator: Andreu Catala (UPC). UdG: Joan Colomer.
Partners: UPC / CETENASA / CNRS / UdG.
Dates: 31/Jan/2001 - 30/Sep/2002.
Sponsor: DURSI (Generalitat de Catalunya) - CTP.
Reference: CTP-ITT 2000-18.
Description: Obtenció de descripcions qualitatives de patrons temporals per l'identificació en línia de situacions de funcionament anormal.
Resolució 16/01/2001

Knowledge Discovery Network.
Principal investigator: Joaquim Meléndez.
Dates: 01/Jan/2001 - 01/Dec/2004.
Sponsor: European Union.
Reference: (IST-2001-33086).
Description: The KDNet (= Knowledge Discovery Network of Excellence) is an open Network of participants from science, industry and the public sector. The major purpose of this international project is to integrate real-life business problems into research discussions and to collaborate in shaping the future of Knowledge Discovery and Data Mining. The project is funded by the European Commission under IST-2001-33086 and has started in February 2002.

Project responsible is Dr. Michael May from Fraunhofer Institute for Autonomous Intelligent Systems (AIS/KD) in Sankt Augustin, Germany.

Encapsulado de microsistemas en formato MCM-D.
Principal investigator: Enric Cabruja Casas (CNM).
Partners: CNM-CSIC, UdG.
Dates: 01/Feb/2000 - 01/Feb/2002.
Sponsor: CICYT.
Reference: TIC99-0846.
Description: ---

Controle avance pour des procedes de depollution.
Principal investigator: Josep Aguilar/Josep LLuis de la Rosa.
Partners: CETENASA, UPC, CNRS_LAAS, UdG, UIPerpignan.
Dates: 04/Feb/1998 - 04/Feb/2000.
Sponsor: DGR (Generalitat de Catalunya).
Reference: ITT-CTP97-9.
Description: ---

Aplicación de los módulos Multichip de tipo D a Sistemas de Potencia.
Principal investigator: Enric Cabruja (CNM).
Partners: CNM-CSIC, UdG.
Dates: 01/Nov/1997 - 01/Nov/2000.
Sponsor: CICYT.
Reference: TIC97-1160.
Description: Este proyecto desarrollará una primera aproximación al encapsulado de sistemas de potencia con tecnología de módulos multichip. Es conocido el amplio campo de aplicación de las tecnologías multichip para dispositivos y sistemas de baja señal. Las ventajas que dichos módulos presentan respecto al encapsulado tradicional son una mayor densidad de integración, una mayor velocidad de operación, una disminución de los costes de producción y montaje de los dispositivos finales. Concretamente se van a desarrollar las técnicas de encapsulado necesarias para la integración de circuitos de media potencia dentro de un módulo multichip tipo D. El dispositivo final fabricado se va a montar com controlador autónomo de tracción del robot autónomo ROGER de la Universitat de Girona. El módulo multichip se integrará en una placa de PCB estándard junto con la circuitería adicional necesaria para el algoritmo de control y comunicación con el microprocesador del robot.