Joint Attention for a Kind of Facial Expression Robots with the Same Structure

In order to improve the abilities of social interaction and cognition in human-robot cooperation between the user and the robot, a static vision based joint attention skill is implemented for a kind of facial expression robots with the same structure. Firstly, the image pixel coordinate of the user's attention focus and the distance between the focus and the video camera are acquired. Secondly, world coordinate of the user's attention focus is solved with coordinate transformation technology. Thirdly, inverse kinematics of the robot body is computed based on geometry method. Fourthly, servo motors are controlled with joint angle acquired above to realize human-robot joint attention. Some experiments are carried out on simulation and physical platform at last to verify the effectiveness and accuracy of the proposed method. It provides a solution for realizing joint attention skill in human-robot interaction and cooperation.