A PCA-based Method to Map Aesthetic Movements from Dancer to Robotic Arm

Abstract

The research field of motion is shared by both robotics and dance for both their respective pragmatic and aesthetic purposes. Dance is an art form which uses the graceful motion of the performer's body to create an expression, while robotics focuses on the motion control of articulated bodies. We propose an approach to combine the graceful motion of a dancer and the robotic motion of an articulated arm to create a unique choreography. Our mapping approach is based on the results of a principal component analysis of dancer motion, projected onto the degrees of freedom of the robotic arm. The challenge here is to find a kinematic relationship between two distinct kinematic structures, a dancer's body and a robotic arm, to create an enhanced aesthetic experience. Various joint assignment policies will be examined and compared to identify the characteristics that optimize the audience's opinion of the robot's movement aesthetics and the appeal of the human-robot choreography.