A Multimodal Tactile Sensor for Material Recognition, Force Measurement, and Slip Detection in Robotic Gripper

Abstract

This article presents a compact multimodal tactile sensor integrating triboelectric, bending, and pressure-sensing elements for robotic automation manipulation. Four triboelectric electrode pairs with different friction materials are used for surface-material recognition, while a bending sensor embedded in a centrally located silicone protrusion responds to tangential deformation associated with object slip. A pressure sensor beneath a rigid support structure measures the normal gripping force. Finite-element analysis was used to optimize the geometry of the central protrusion and the support-pillar height to improve the sensitivity and stability of slip detection. Experimental results show that the pressure-sensing unit provides a response range of 0–100 N, whereas the bending-sensing unit responds to tangential displacements of 0–4 mm. The triboelectric-sensing unit achieves classification of six materials via a 1-D convolutional neural network, with a test accuracy of 99.17%. In a robotic-grasping experiment, the sensor recognized the surface materials of three cups and detected slip induced by a progressively increasing payload. The detected slip signal automatically triggered an increase in gripping force and prevented further downward motion of the cup. These results demonstrate the feasibility of integrating material recognition, force measurement, and slip detection into a single tactile sensing device for adaptive robotic grasping.