Human-Manipulator interface using hybrid sensors via CMAC for dual robots

Abstract

This paper presents a novel method that allows a human operator to communicate his motion to dual robot manipulators by performing his double handarms movements, which would naturally carry out an object manipulation task. The proposed method uses hybrid sensors to obtain the position and orientation of the human hands. Although the position and the orientation of the human hands can be obtained from the sensors, the measurement errors increase over time due to the noise of the devices and the tracking error. A cerebellar model articulation controller (CMAC) is used to estimate the position and orientation of the human hands. Due to the limitations of the perceptive and the motor, human operator can not accomplish the high precise manipulation without any assistants. An adaptive multi-space transformation (AMT) is employed to assist the operator to improve the accuracy and reliability in determining the posture of the manipulator. With making full use of the human hand-arms motion, the operator would feel kind of immersive. Using this human-robot interface, the object manipulation task done in collaboration by dual robots could be carried out flexibly through preferring the double hand-arms motion by one operator.