Maha Salem, Stefan Kopp, Katharina Rohlfing, Frank Joublin, "Investigating the Effect of Robot Gesture", IEEE Int. Symposium on Robot and Human Interactive Communication (RO-MAN), 2011.

The generation of communicative, speech-accompanying robot gesture together with an evaluation of the effects of such multimodal behaviour is still largely unexplored. The main objective of the presented study is to shed light onto human perception and understanding of gestural machine behaviors and how these can be used to design more natural commu- nication in social robots. The approach is twofold: Firstly, a speech-gesture generation model on a robotic platform has been implemented. The aim is to enable the Honda humanoid robot to flexibly produce synthetic speech and expressive hand gesture from conceptual representations and planning, while not being limited to a predefined repertoire of motor action. Secondly, the achieved flexibility in robot gesture has been exploited for controlled experiments. These center around how humans perceive various gestural patterns performed by the humanoid robot as they interact in a situational context.