The main objective of POETICON++ is the development of a computational mechanism for generalisation of motor programs and visual experiences for robots. To this end, it will integrate natural language and visual action/object recognition tools with motor skills and learning abilities, in the iCub humanoid. Tools and skills will engage in a cognitive dialogue for novel action generalisation and creativity experiments in two scenarios of "everyday activities", comprising
- behaviour generation through verbal instruction, and
- visual scene understanding.
- Vadim Tikhanoff
- Panagiotis Dimitrakis
- George Karakatsiotis
- Giovanni Saponaro
- Duarte Aragão
Behaviour generation through verbal instruction
Through a number of object and action learning sessions, the iCub will acquire knowledge of specific objects and actions. These will be stored in the PRAXICON-episodic memory (i.e. visual, motoric, tactile representations as provided by iCub will be stored in the PRAXICON-episodic, along with language labels that will be given to those objects and actions). They will also be linked automatically to the PRAXICON-Semantic Memory, which will allow for use of this knowledge in a systematic way for generalisation.
Visual scene understanding
Interpretation and understanding of human action.