CozmoNAOts: Designing an Autonomous Learning Task with Social and Educational Robots

Abstract

Learning tasks designed with social or educational robots are becoming increasingly commonplace, with emerging methodologies being proposed regarding the design of interaction patterns for child-robot-interaction (cHRI). Yet, technological limitations remain a strong barrier to the implementation of fully autonomous robot tutoring systems. In addition, there is currently no research on how social and educational robots might be combined when designing learning tasks. Consequently, in this paper we describe the design and pilot testing of a (semi) autonomous learning task designed with a social robot (NAO) and educational robot (Cozmo) targeted at children’s computational thinking skills. Preliminary data from a pilot phase with 53 children is promising, and results are discussed with regards to identified challenges. Several solutions are proposed in the context of designing interactions for cHRI.