High-Performance Vision-Based Tactile Sensing Enhanced by Microstructures and Lightweight CNN

Abstract

Tactile sensing is critical in advanced interactive systems by emulating the human sense of touch to detect stimuli. Vision-based tactile sensors are promising for providing multimodal capabilities and high robustness, yet existing technologies still have limitations in sensitivity, spatial resolution, and high computational demands of deep learning-based image processing. This paper presents a comprehensive approach combining a novel microstructure-based sensor design and efficient image processing, demonstrating that carefully engineered microstructures can significantly enhance performance while reducing computational load. Without traditional tracking markers, our sensor incorporates an surface with micromachined trenches, as an example of microstructures, which modulate light transmission and amplify the response to applied force. The amplified image features can be extracted by a ultra lightweight convolutional neural network to accurately inferring contact location, displacement, and applied force with high precision. Through theoretical analysis, we demonstrated that the micro trenches significantly amplified the visual effects of shape distortion. Using only a commercial webcam, the sensor system effectively detected forces below 5 mN, and achieved a millimetre-level single-point spatial resolution. Using a model with only one convolutional layer, a mean absolute error below 0.05 mm was achieved. Its soft sensor body allows seamless integration with soft robots, while its immunity to electrical crosstalk and interference guarantees reliability in complex human-machine environments.