PLANNING & LEARNING RESEARCH GROUP Universidad Carlos III de Madrid
Current Projects
HAPE (2022)
Human-Aware Planning & Execution
Financing: PID2021-127647NB-C21, financiado por el Ministerio de Ciencia e Innovación MCIN/AEI/10.13039/501100011033/ y “FEDER Una manera de hacer Europa”.
Description: As Artificial Intelligence (AI) technologies enter our everyday lives at an increasing pace there is a greater need for AI systems that work synergistically with humans. The main objective of this project is to analyze, develop and evaluate new methods, algorithms and tools that support the use and application of Automated Planning and Machine Learning to create intelligent systems in real scenarios where humans play a main role due to its implication on the design of the system, the deliberation or reasoning processes or the actual resolution of the specific task. We turn our attention towards human-centric applications for: (1) supporting the human in the definition of use cases, (2) learning from human and agent deliberation to improve automated problem solving and (3) learning from plan observations to support the human in reaching her goals.
Contributors:
Universidad Carlos III de Madrid (PLG), GPRS AI Group
Universidad Politécnica de Valencia
Financing: This work has been partially funded by FEDER/Ministerio de Ciencia e Innovación - Agencia Estatal de Investigación/RTC2019-007377-4
Description: Software que garantice una planificación estratégica del transporte público multimodal y urbano completamente adaptado a las necesidades de movilidad de las Smart Cities, capaz de competir con el uso del transporte privado. Para ello, se generarán matrices origen-destino construida a partir de los datos de movilidad de operadores de telefonía utilizando técnicas de AI (Artificial Intelligence). Asimismo, se desarrollará un motor de razonamiento que contará con nuevos algoritmos de optimización teniendo en cuenta las restricciones para resolver el problema de planificación de la red de tránsito denominado por sus siglas en inglés, TNP (Transit Network Problem).
Contributors:
Universidad Carlos III de Madrid (PLG), Goal Systems
Resolución de Casos de Uso de Aprendizaje por Refuerzo (2019)
Financing: This project is granted by Repsol
Description:
Contributors:
Universidad Carlos III de Madrid (PLG), Repsol
Arquitecturas para la Capacitación Social Basadas en Planificación Automática (2019)
Financing: This project is granted by the Spanish Goverment (Ministerio de Economia y Empresa) and FEDER, UE funds under project TIN2017-88476-C2-2-R
Description: Providing agents (both human and software) with recommendations on future behavior (actions, plans or goals) based on learning models of their past behavior and recognition of their present actions, plans or goals. In order to achieve the objective, we will integrate techniques from machine learning, activity/plan recognition and automated planning.
The key idea is to analyze information from sensors in order to generate diverse models of an agent's behavior, such as planning domain models, probabilistic models, or generic classifiers.
Contributors:
Universidad Carlos III de Madrid (PLG), GPRS AI Group Universidad Politecnica de Valencia
Former Projects
Click to expand each project
Learning Similarity Metrics Between Simulation and the Real World (2020)
Description: Using knowledge gathered from simulated environments in the real world can be seen as a transfer learning approach where simulated environments are the source tasks and the real world is the target task. The accuracy of the transfer depends on how similar the source simulations and the real world are. From the Reinforcement Learning point of view, that transfer can be understood as how useful the policies learned in the simulations are for solving the problems of the real world. Probabilistic Policy Reuse (PPR) raised as a transfer learning method that was able to accurately reuse policies learned in source tasks (in this case, simulation) in a target task (the real wold). In addition, one of the main characteristics of PPR is that it was able to de ne a distance metric between the di erent tasks. In this project, we plan to extend PPR concepts to fi nd accurate similarity metrics between simulation and the real world.
Contributors:
Universidad Carlos III de Madrid (PLG), JP Morgan
Planificación de trabajos (2020)
Financing: This project is granted by Indra Soluciones Tecnologías de la Información, S.L.U.
Description: -
Contributors:
Universidad Carlos III de Madrid (PLG), Indra Soluciones Tecnologías de la Información, S.L.U.
Autonomous Social Robotics for Pediatric Assistance and Active Aging (2019)
Financing: This project is granted by the Spanish Goverment (MINECO. RTC-2017-6753-4) in collaboration with Goal Systems
Description: Software de planificación automática y optimizada que garantice la explotación eficiente de la capacidad de tráfico de trenes en estaciones de alta congestión. La herramienta se encargará de buscar un aprovechamiento cercano al 100% de la capacidad ofrecida por cada estación teniendo en cuenta la complejidad de la topología de la red de las estaciones, las reglas de transporte ferroviario y los servicios que se deben considerar
Contributors:
Universidad Carlos III de Madrid (PLG), Goal Systems
Financing: This project is granted by the Spanish Goverment (MINECO. RTC-2016-5407-4) in collaboration with Goal Systems
Description: Herramienta software de planificación óptima que resuelva problemas relacionados con la generación de rutas.
Uno de los outputs de la aplicación será la generación óptima y automática de oferta de rutas teniendo como entrada una demanda.
Esta demanda puede ser de recogida de pasajeros/material desde múltiples orígenes hacia un mismo punto destino, o bien, de reparto de pasajeros/material desde un mismo punto origen a diversos destinos.
La aplicación servirá para cualquier sector del transporte colectivo, validando su performance en el transporte escolar.
Contributors:
Universidad Carlos III de Madrid (PLG), Goal Systems
Financing: This project is granted by the European Space Agency (4000117648/16/NL/GLC/fk), subcontract with GMV (GMVAD GOTCHA/UC3M 30809/16)
Description:
GOAC TRL Increase Convenience Enhancements Hardening and Application Extension (GOTCHA). The project combines classical temporal planning and timeline-based planning for the autonomous control of a space rover, the ESA RAT platform. The aim is to achieve an autonomous framework for planetary-exploration rovers, increasing their Technology Readiness Level (TRL) for use in future space systems such as the upcoming Mars Sample Return mission. Field trials in Colmenar Viejo have shown the developed system fulfills all the requirements
Contributors:
Universidad Carlos III de Madrid (PLG), GMV
Financing: This project is granted by the Spanish Goverment (MINECO) and FEDER,UE funds
Description: Smart homes, when applied to elder people, can be considered as residential houses equipped with sensors and automated devices whose goal is to deliver care and monitoring of the people living in them. Such assistance must be defined in the long-term, and it must attempt to balance the immediate specific needs of the user with the long-term effects that the robot’s and assistance technologies can potentially have on the user’s developmental trajectory. The LifeBots project will investigate the use of lifelong technologies as a key resource to adapt to new goals, user’s preferences or environmental issues that require a tailoring of current assistive settings.
Contributors:
Universidad Carlos III de Madrid (PLG), Universidad de Extremadura (Robolab), Universidad de Málaga (ISIS), Universidad de Jaén (M2P), Universidad de Castilla La Mancha (SIMD)
Financing: This project has received funding from the European Union's Seventh Framework Programme for research, technological development and demonstration under grant agreement no 601116
Description: The project is focus on robotized Comprehensive Geriatric Assessment. In particular, it focuses on the development of CLARC, a mobile robot able to receive the patient and his family, accompany them to the medical consulting room and, once there, help the physician to capture and manage their data during the Comprehensive Geriatric Assessment (CGA) procedures.
Contributors: Universidad de Málaga (UMA), Universidad Carlos III de Madrid (UC3M), Metralabs Mobile Robots, Université de Technologie de Troyes (UTT), Servicio Andaluz de Salud (SAS)
Description: The main objective of the project is to analyze the problem of goal management for long-term autonomous systems, design appropriate algorithms for addressing the different components of goal management, and develop software tools that help on the application of this technology to Smart Cities tasks. The long-term goal of fully autonomous systems is a highly ambitious objective for many fields, including Artificial Intelligence or Robotics. Thus, we focus on one of the aspects that has been less explored in the literature, but with a big impact on the overall objective of long-term autonomy: automatic goal management.
Contributors: GPRS AI Group Universidad Politecnica de Valencia, Universidad Carlos III de Madrid
Description: The objective of this proposal is the development of THERAPIST, a socially interactive robot for neuro-rehabilitation assistance
Financing: MICINN (Spanish Department for Science)
Contributors: Robolab Universidad de Extremadura, ISIS Universidad de Málaga, Hospital Universitario Virgen del Rocío, M2P Universidad de Jaén, Universidad Carlos III de Madrid
Planinteraction (2012)
Project number: TIN2011-27652-C03-02
Description: Multi-agent interaction for planning
Financing: MICINN (Spanish Department for Science)
Contributors: Universidad de Granada (UGR), and Universidad Politécnica de Valencia (UPV) on the development of multi-agent planning systems
Adapta (2013)
Description: The main objective of this project is to develop and integrate technological solutions in order to implement new forms of customization and interaction in the field of digital content. It aims to develop innovative mechanisms for user-advertisement interaction. In this sense, one of the goals, where the group PLG is working at present, is the human-robot interaction.
Financing: --
Contributors: INDRA Sistemas, INDRA Software Labs, Universidad de Málaga, Universidad de Cáceres, Universidad Carlos III de Madrid
Space Situational Awareness (2011)
Description: DC-II Prototype Tasking & Data Centers WP 2. Sensor Planning Services. On the development of a tool for planning sensors observations.
Description: Publishing Technology Platform, Disclosure and e-Book Distribution to Promote Reading and Literary Creation in Castilian through Social Networks..
Financing: --
Contributors: Digimate Computer S.L. and Lakotel Soluciones S.L.L.
Description: Data Analysis and Learning autonomous behaviors for systems generation decision support in business simulator.
Financing: Simuladores Empresariales S.L. y proyecto de Investigación Fundamental Orientada a la Transmisión de Conocimiento a la Empresa (TRACE) del MICINN .
Contributors: -
AIRPP (2011)
Description: Robotics Instruction Learning by Policy and Planning
Financing: Comunidad de Madrid (regional government of Madrid)
Contributors: -
AIRPP (2011)
Description: Robotics Instruction Learning by Policy and Planning
Financing: Comunidad de Madrid (regional government of Madrid)
Description: Research project on Briding Courses for Mathematics.
Financing: -
Contributors: German Research Centre for Artificial Intelligence, University of Vienna, University of Saarland, Tampere University of Technology, Open Universiteit Nederlands, Eötvös Loránd University Budapest, Universität Kassel, Université Montpellier 2
Description: Planning Execution and Learning Architecture. The goal is to build a domain-independent architecture for planning, execution and learning in domains as robotics, or video games.
Financing: MICIIN
Contributors: Universidad de Granada and Universidad Politécnica de Valencia.
CCI (2009)
Description: Our task consists on the use of Bayesian belief networks for cost estimation
Financing: ESA
Contributors: Coordinated by GMV and in collaboration with Skysoft
Ericsson: MOLE y AUKB (2009)
Description: Research projects on the application of planning to data mining.
Financing:Ericsson
Contributors: -
RAPLAN (2009)
Description: Recognition based on Planning Activities. Research project on planning for activity recognition.
Financing: CAM (regional government of Madrid) and UC3M
Contributors:
TIMI (2008)
Description: Our task consists on the development of planning technology for logistics applications.
Financing: CDTI
Contributors: Coordinated by Atos Origin and in collaboration with Acciona Transmediterránea
IPLA (2007)
Description: Research project on integration of planning and machine learning.
Financing: CAM (regional government of Madrid) and UC3M.
Contributors: -
MiPlan (2006)
Description: Joint Initiative for automatic task planning. Research project on mixed-initiative planning and machine learning.
Financing: CAM (regional government of Madrid) and UC3M
Description: Research on user-oriented adaptive planning systems applied to education
Financing: MEC (Spanish Department for Education)
Contributors: Universidad de Granada (UGR), de Valencia (UPV), de Educación a Distancia (UNED), de Girona, and de Castilla la Mancha
SAMAP (2002)
Description: Research project on adaptive multi-agent planning systems context dependent.
Financing: MCyT (Spanish Department for Science and Technology)
Contributors: Universidad de Granada (UGR), de Valencia (UPV), de Educación a Distancia (UNED), and the IIIA institute of the Spanish Research Council (IIIA-CSIC)
PLANET II (2001)
Description: ESPRIT research Network of Excellence on Planning
Financing:
Contributors:
Diseño automático de programas de control secuencial: aplicación al control de procesos térmicos en la industria láctea (1999)
Description: Automatic design of control programs for dairy industry
Financing:CICYT (Spanish Department for Research)
Contributors: Universidad de Granada, PULEVA, and ABB
SHAMASH (1999)
Description: This project that applies AI techniques to process modelling, simulation and optimization.
Financing:
Contributors:
Cost-282 (-)
Description: The primary objective of the Action is to develop and implement computerized systems for extracting previously unknown, non-trivial, and potentially useful knowledge from structurally complex, high-volume, distributed, and fast-changing scientific and R&D databases within the context of current and newly developing global computing and data infrastructures such as the GRID.
Financing:-
Contributors: -
RNPST (-)
Description: National Network Planning, Sequencing and Temporal Reasoning
Financing: -
Contributors: -
Red Nacional de AgenCities.ES (-)
Description: Creating an innovative environment for communication of intelligent agents. National network of kind place in Agent Cities.