Enhancing Vision-Based Tactile Sensing Through a Fingernail-Inspired Structure for Improved Contact and Texture Detection

Vision-based tactile sensors (VBTS) have demonstrated strong potential for high-resolution contact measurement, yet their structural design offers room for further improvement. In this study, we examine the impact of incorporating a human fingernail-inspired structure on the performance of VBTS in measuring contact areas and detecting fine textures. Tensile tests were conducted on four flexible skin materials, and VBTS units were fabricated both with the nail structure (w-VBTS) and without it (wo-VBTS) to evaluate contact areas on flat surfaces. Based on the tensile test results regarding durability and material transparency, the optimal skin material was selected for the w-VBTS, which was then used in fine step detection experiments. Our findings reveal that the nail structure effectively reduced excessive skin deformation and increased the contact area. Furthermore, fine texture detection at the nail tip was achieved with a maximum error of 0.69 mm. These results suggest that emulating the structural features of the human finger can enhance the performance of VBTS in robotic manipulation tasks.