Development of an emotional expression platform based on LMA-shape and interactive evolution computation

摘要

Development of systems for Human-Robot Interaction (HRI) has been a subject of intense interest to many researchers in recent years. Communication, reflecting intent or emotional state, is a central issue in such systems. This paper propose the use of bodily movement for robotic expression of emotional states. To realize such expression a robot must be equipped with a set of articulators for head, trunk and arms. Normal human movements have been taken as a model. Observations of human upper-body motions were recorded by a depth camera. The visual data thus obtained was analyzed by means of the Laban Movement Analysis (LMA) method to elicit critical parameters for generating corresponding robotic movement. Experiments were conducted with the upper-body of a Seamless robot to verify and evaluate the effectiveness of the proposed system.