Search

Tactile sensors are widely used in electronic skin, medical robotics, and other fields, which is a crucial foundation for the future development of artificial intelligence. Their primary function is the accurate measurement of force in various modalities, including temperature, humidity, static, and dynamic conditions. Therefore, the calibration of tactile sensors has become highly important. The traditional force sensor calibration techniques fail to meet the calibration requirements of multimodal tactile sensors. In this article, a high-precision multimodal calibration method and device for tactile sensors are proposed. The device consists of a temperature and humidity chamber, a six-axis robotic arm, voice coil motors, piezoelectric pressure sensors, control circuits, and a data processing system. The output’s high-precision force based on the electromagnetic control of the voice coil motor uses a six-axis manipulator to realize the calibration of any position in space and designs a control algorithm to realize the static and dynamic calibration. The experimental results demonstrate that the device can achieve calibration of force ranging from 100 mN to 10 N at dynamic frequencies of 0.1–10 Hz within a temperature range of 0 °C–60 °C and a humidity range of 30%–60%RH. The maximum relative standard uncertainty is $\le2.5$ %.

Although vision-based defect detection has been widely applied in the industry, vision information alone is still insufficient for certain flexible materials such as leather, as their quality control may involve haptic properties. This article introduces a solution based on the fusion of vision and tactile data to evaluate the overall quality of flexible materials, such as leather. Firstly, a prototype equipment for capturing visual and tactile data is introduced. Specifically, the tactile data consists of XYZ-axis three components and is interpolated and visualized to align with the visual data. The aligned tactile data is then combined with the visual data and the derived defect segmentation map to form a cross-modal dataset. Additionally, a DeepLabv3+ derived network, which can be considered as a two-stream network with cross attention between visual and tactile data, is proposed to detect surface and non-surface defects. Experiments demonstrate that the proposed network exhibits a significant improvement of 10% compared to the vision-only network. The ablation study also examines the performance comparison of different fusion strategies. Furthermore, a grading strategy that integrates defect detection with haptic properties is finally introduced to assess the quality of flexible materials.

With the development of high-power laser technology, accurate measurement of laser power has been a difficult task. The light pressure weighing method (LPWM) can replace the calorimetric laser power meter as a standard measurement device in the high power range due to its fast response, high accuracy, and direct traceability to mass. When the laser power is less than 1 kW, the accurate measurement of light pressure requires a weighing sensor with a higher level of resolution, and multiple environmental disturbances will significantly affect the measurements. We present a high-precision LPWM power meter based on a W-shaped dual-reflection structure and a 1- $\mu \text{g}$ resolution weighing sensor, corresponding to a laser power measurement resolution of 0.78 W. Simulation of air parameters near the sensing mirror shows that the laser causes different air pressure changes at the top and bottom of the sensing mirror, and monitoring of temperature, relative humidity and laser power all show linear trends. By using the segmented compensation strategy, we successfully achieve more stable measurements and significantly improve the level of measurement repeatability. The expanded uncertainty after compensation is better than 2% for laser power above 150 W and better than 1% for power above 500 W. This article makes it possible to quickly and accurately measure laser power from 150 W to 1 kW in conventional environments and is also expected to replace calorimetry as the standard device in the sub-kilowatt range.