RoboTAskProject reference: 624424
Funded under :
Action words Learning in a Humanoid Robot by Discovering Tool Affordances via Statistical Inference
Total cost:EUR 179 739,6
EU contribution:EUR 179 739,6
Topic(s):FP7-PEOPLE-2013-IEF - Marie-Curie Action: "Intra-European fellowships for career development"
Call for proposal:FP7-PEOPLE-2013-IEFSee other projects for this call
Funding scheme:MC-IEF - Intra-European Fellowships (IEF)
Recently, humanoids have started to be employed in linguistic investigations with the dual aim of endowing robots with the capability to communicate and interact with humans, and further understanding the mechanisms underlying language development. The proposed research aims at creating a developmental cognitive model for the iCub humanoid robot for grounding the meaning of language in tool affordances. Despite it is firmly clear that language has to be grounded in sensorimotor experience, recently it has became increasingly evident the importance of going beyond simple sensorimotor grounding. To this end, statistical inference provides an original and innovative methodology that can serve in grounded theories of meaning. Computational modeling, based on statistical inference over hierarchies of flexibly structured representations, can address important problems related to human intelligence and cognition, such as the learning of language and causal relations. This research will permit to endow the iCub robot with the ability to understand language during the interaction with a human tutor and provide an experimental framework for investigating how language interacts with other cognitive processes, such as motor control. Additionally, a human-robot interaction study will permit to analyze how people experience interaction through language with the iCub robot. On the one hand, the proposed artificial cognitive system can progress theories of language learning in robots, with consequent advances in the design of human-robot communication systems, which can lead to a new generation of interactive robots. On the other hand, experiments performed on the model, by generating new predictions, can have important implications in cognitive science. The grant of the fellowship will provide the Researcher with the opportunity to further develop interdisciplinary skills in developmental robotics methods, and transferable skills for R&D in academia and service robotics industry.
EU contribution: EUR 179 739,6
VIA MOREGO 30
Tel.: +39 01071781565
Fax: +39 0107170817