Abstract—This paper describes a new compact variable stiffness joint mechanism. This mechanism is so small as to be embedded into the finger joint of a human-sized robotic hand. In the mechanism, a silicone rubber cushion, which is inserted between the motor and the output link is effectively used to adjust the joint stiffness and to sense external forces. The joint stiffness can be controlled by restricting the degree of cushion deformation with a cushion support wall (CS-wall) actuated by a pneumatic system. However, the joint stiffness is adjusted to only the lowest or highest level, because of the difficulties of achieving precise position control of the CS-wall by the pneumatic system. Conversely, the detection of external forces shows good performance. The change in the pressure inside the cushion provides information on the external forces applied to the link. We applied this function to measure the friction force between the robotic finger and the object, and attempted to determine differences in the friction force among twelve types of objects. The experimental results show that the mechanism is able to determine differences in the friction force and its oscillation between the objects. We confirmed that this ability can be used to classify objects.

Index Terms—variable stiffness joint, robot finger, force estimation, estimation of contact situation

Cite: Shinya Kajikawa and Aruto Suzuki, "Compact Variable Stiffness Joint with Force Sensing Ability-Application to Classify Contact Surfaces," Jounal of Automation and Control Engineering, Vol. 3, No. 3, pp. 215-221, June, 2015. doi: 10.12720/joace.3.3.215-221