Description du projet
Un nouveau réseau pour déployer davantage de robots sur le terrain
L’exploitation minière, l’agriculture, l’exploitation forestière et la construction sont les principaux domaines d’application des robots de terrain, qui peuvent fonctionner au sol, sous l’eau, dans l’air ou dans l’espace. Dans les fermes, on trouve des robots conçus pour des tâches spécifiques telles que la cueillette des fraises et la traite des vaches. Dans les mines, on utilise des robots pour cartographier les passages inondés et analyser les concentrations de minéraux. Le projet DEEPFIELD, financé par l’UE, a pour objectif de promouvoir l’apprentissage profond dans le domaine de la robotique de terrain. Il aidera l’INESC TEC (Institut pour l’ingénierie informatique et des systèmes, les technologies et les sciences), au Portugal, à devenir un centre d’excellence en robotique de terrain en établissant des relations avec des instituts de recherche européens de premier plan.
Objectif
Robots are active agents that need to interact with the physical world, to do so, robots are equipped with different sensors, whose data is used to build models that ultimately will allow robots to plan actions and make decisions.
Currently, there is strong focus in developing deep learning strategies “data driven” to help solve this perception problem, even though these approaches work well in dataset and benchmark scenarios. There are still strong limitations in the use of this techniques in real world robot activities, specially due to the strong dynamics in robots operational environment, that is pushing the development of new tools and methods to make these approaches feasible in the real world.
INESC TEC is strongly committed to become a centre of excellence with focus on field robotics, in particular, in the aerial and underwater robotics domain. In the last years, the centre for Robotics and Autonomous systems, of INESC TEC has advance its scientific knowledge in sensing and perception methods for robots navigation and localization in harsh operational environments. The key objective of INESC TEC is to become one of the European references in field robotics, and help to bring robot technology to solve real life problems where human intervention is still limited or non-existent.
This proposal aims at creating solid knowledge and productive links in the global field of deep learning in field robotics between INESC TEC and established leading research European institutions, capable of enhancing the scientific and technological capacity of INESC TEC and linked institutions (as well as the capacity of partnering institutions involved in the twinning action), helping raising its staff’s research profile and its recognition as an European research centre of excellence in field robotics. In particular, it takes INESC TEC and places it as the pivot of a network of excellence, involving four international leaders in deep learning technology and fied robotics.
Champ scientifique
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensors
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringrobotics
Programme(s)
- H2020-EU.4.b. - Twinning of research institutions Main Programme
Thème(s)
Régime de financement
CSA - Coordination and support actionCoordinateur
4200 465 Porto
Portugal