A Compact Visuo-tactile Robotic Skin for Micron-level Tactile Perception

Visuo-tactile skins (VTS) are classified into coated-type, marker-type and thermochromic-type. Coating wear resistance and spatial resolution are central issues for the coated-type VTS. This paper proposes a wire-drawing process applied to the VTS to enhance the coating robustness. The wire-drawing coating has more stable mechanical properties, including high adhesion, adaptive deformation, and rupture resistance. Under different degrees of deformation, it can remain continuous and undamaged. In addition, the wire-drawing coating has a nanoscale particle distribution, whose microstructures are similar to the spraying coating to provide micron-scale spatial resolution. Via the wear resistance test, the wire-drawing coating has higher wear resistance than the spraying coating. Further, the VTS combined with a multitask learning model can identify textures and clothing properties of 18 different fabrics with more than 98% recognition accuracy. The results show that the VTS using the new process can provide highly learnable tactile data.