Incipient slip detection method for soft objects with vision-based tactile sensor

Abstract

Incipient slip detection plays an important role in human and robotic grasping. Recently, vision-based tactile sensors are increasingly used to detect incipient slip because of their high spatial resolution. However, current methods can only be applied to rigid objects and do not work as well on soft objects. To fill this gap, an incipient slip detection method which can be applied to arbitrarily flexible objects with vision-based tactile sensors is proposed. This paper firstly describes the generalized model of slip calculation with vision-based tactile sensors. Then an approximate linear relation between the derivative of object and sensor deformation field is demonstrated, leading to an incipient slip detection method based on the distributed force and deformation gradient, whose results can indicate the degree of incipient slippage. The accuracy and applicability of this method are verified through finite element analysis and real grasping experiments of some common objects with a robot hand.