ANIMATAS

Advancing intuitive human-machine interaction with human-like social capabilities for education in schools

ISIR

LOGO_SU_HORIZ_SIGNATURE_CMJN_JPEG

 

Europe Flag

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 765955

Identifiez-vous pour accéder à la partie privée du site ou le modifier. Déconnexion.

ESR5 - Socially compliant behaviour modelling for artificial systems and small groups of teachers and learners

Early-Stage Researcher: Sahba Zojaji

Sahba Zojaji is a Marie-Curie PhD fellow of Computer Science at the KTH University, Sweden. He received his M.Sc. in Computer Engineering-Artificial Intelligence & Robotics from Iran University of Science & Technology (IUST) in 2011. After his ten-year career as an engineer and researcher in different companies and universities in Iran, eventually Sahba joined the LITIS laboratory in 2016, where he accepted a research engineer position at INSA of Rouen Normandy, France. 

His research interests include affective computing, cognitive modelling, Human-Computer Interaction, humanoid and social agents. He has focused on Embodied Conversational Agents during the last two years. He currently resides in Rouen with his wife and his daughter, however, he is going to move to Stockholm in the near future.

 


Main host institution: KTH

Supervisor: Christopher Peters (KTH), in association with Mohamed Chetouani, Catherine Pelachaud (UPMC) and Chloé Clavel (IMT)

Secondment institution: IMT; UPMC; SBR

Objectives: While many modern education scenarios involving robots and agents involve the participants (teachers and children) moving to and accommodating the robot, a vision for future training scenarios is that robots will be capable of moving to groups of learners and teachers in order to accommodate them. This requires systems that are capable of not only navigating, but also positioning and introducing themselves appropriately in order to engage in interaction i.e. they must be socially compliant. For example, artificial systems should join a group in a suitable manner, minimising intrusions to a group that is engaged in an ongoing pedagogical scenario, and do so by finding a suitable position for themselves in a formation of conversation partners, maintaining appropriate social and sensory distances. The central aspect in this ESR project will be to develop socially compliant small group behaviours for robots and agents tailored to small groups of learners and teachers. This will be accomplished through procedural models and machine-learning approaches based on data corpora from human-human and human-machine interactions and through the use of experiments involving human participants and groups of virtual replicas in virtual reality. 

Expected results: Completed draft of PhD dissertation, algorithms and tools for modelling compliant behaviours between mobile systems and small groups, peer-reviewed publications, international journal and conference publications 

 

For further information, contact: Christopher Peters

chpeters@kth.se


ANIMATAS – MSCA – ITN – 2017 - 765955 2