3-D Interface for Humanoid Robot Operation

Abstract

The study of human body motions, especially upper and lower limbs motions, help to understand how the human body works. This paper presents a method to obtain the human limb angles through a kinematic model depicted by Roll-Pitch-Yaw rotation matrix and subsequently control the humanoid robot motions. The main advantage of this model is the detailed representation of each joint motion expressed in the coordinate axes (x, y, z). The estimation of human limb angles is performed with information obtained by algorithms of artificial vision and artificial intelligence. In order to reduce the latency between the human motion capture and robot motions, a fuzzy logic controller is implemented to control each robot joint due the less complexity of these kind of algorithms. The final robot limbs angles are compared with the human limbs angles to obtain a final error between those measurements. This method shows an improvement of 5% on the final robot posture regarding previous works.