A 3D-Printable Dataset for Fair Testing and Comparisons of Tactile Sensors

Existing texture datasets for tactile sensing primarily consist of sensor readings from a specific sensor interacting with available surfaces/objects rather than describing the textures themselves, limiting fair comparison between tactile sensors and hindering reproducible research. In this work, we introduce a six-piece digital dataset of mathematically defined textures designed to be fabricated reliably across different research groups. The dataset consists of surface patterns derived from combinations of sine-waves, giving controlled variation in spatial frequency, amplitude, and directional structure. To evaluate the physical repeatability of this framework, we present a validation methodology applied to the six-textures. We evaluate the reproducibility of these textures across three popular 3D printers and filament types in multiple ways. We first use an RGBD D405 stereoscopic camera to scan the 3D structure to assess physical differences between repeat prints. To see how physical differences affect a typical tactile sensor, we next calculate the difference in images captured using an optical TacTip sensor under controlled contact conditions. Our results show that print quality, particularly peak sharpness and stringing, affects tactile variance, with the Bambu P1P/PLA- and Form 3 resin configurations producing more consistent prints than the Ender-3/PLA+ configuration. To assess potential downstream effects of the variance, we train two types of classifier, random forest classifiers and neural networks, both with PCA preprocessing and data augmentation, to classify textures. By training on one batch of textures, but testing on different unseen batches, we further assessed how similar different batches of prints are. These experiments demonstrate that the Bambu P1P/PLA- and Form 3 resin configurations support both strong within-batch classification and cross-batch generalisation, reflecting the fact that these configurations had low physical and tactile image variance. This work establishes the first openly available, six-texture digital library paired with a metrology pipeline designed to be sensor-independent, providing an open foundation that enables researchers to replicate textures and fairly compare tactile sensing technologies, while our pipeline has been tested on only one tactile sensor (TacTip) and two classifier types, we hope future research will compare many other combinations of sensing technologies.