Comparison Between Admittance and Impedance Control of a Multi-Finger-Arm Robot using the Guaranteed Manipulability Method

In previous studies, we proposed cooperative control algorithms for a finger-arm robot comprising a 3-DOF finger robot and a 6-DOF arm robot. In these algorithms, a finger-arm robot completes a constrained task by integrating admittance and impedance control with manipulability control of the finger. Finger manipulability is controlled by both an approximate global search, i.e., the top search method, and a local optimization method. In this study, the methods that have previously controlled only one finger are extended to control multiple fingers. When the method is extended to multiple fingers, it acquires different characteristics from the case when it is used with only one finger. Therefore, in this paper the differences between the algorithms of these methods are described. Because admittance and impedance control methods offer different characteristics to the manipulability control of the finger, it is necessary to compare these methods from the standpoint of the arm and the finger and organize them for more advanced study. Moreover, when these control methods are adapted for multi-finger robots, it is necessary to consider problems that arise with respect to manipulability and whole motion control of the arm and fingers. This paper compares cooperative admittance and impedance control for a multi-finger-arm robot with manipulability control of the fingers. Experimental results are presented for the simultaneous achievement manipulability, admittance, and impedance control of a multi-finger-arm robot in a threedimensional space. The proposed method demonstrated a strong performance of singularity avoidance, even when a dynamic external force was applied to the multi-finger device. KeywordsMechatronics; Robot; Manipulator; Admittance Control; Impedance Control; Manipulator; Redundancy