Recently, human–robot interaction technology has been considered as a key solution for smart factories. Surface electromyography signals obtained from hand gestures are often used to enable users to control robots through hand gestures. In this paper, we propose a dynamic hand-gesture-based industrial robot control system using the edge AI platform. The proposed system can perform both robot operating-system-based control and edge AI control through an embedded board without requiring an external personal computer. Systems on a mobile edge AI platform must be lightweight, robust, and fast. In the context of a smart factory, classifying a given hand gesture is important for ensuring correct operation. In this study, we collected electromyography signal data from hand gestures and used them to train a convolutional recurrent neural network. The trained classifier model achieved 96% accuracy for 10 gestures in real time. We also verified the universality of the classifier by testing it on 11 different participants.

According to the theoretical equation of the pulse oximeter expressed by the ratio of amplitude (AC) and baseline (DC) obtained from the photoplethysmographic signal of two wavelengths, the difference of the amount of light absorbed depending on the melanin indicating the skin color is canceled by normalizing the AC value to the DC value of each wavelength. Therefore, theoretically, skin color does not affect the accuracy of oxygen saturation measurement. However, if there is a direct path for the light emitting unit to the light receiving unit instead of passing through the human body, the amount of light reflected by the surface of the skin changes depending on the color of the skin. As a result, the amount of crosstalk that varies depending on the skin color affects the ratio of AC to DC, resulting in errors in the calculation of the oxygen saturation value. We made crosstalk sensors and crosstalk-free sensors and performed desaturation experiments with respiratory gas control on subjects with various skin colors to perform oxygen saturation measurements ranging from 60 to 100%. Experimental results showed that there was no difference in the measurement error of oxygen saturation according to skin color in the case of the sensor which prevented crosstalk (−0.8824 ± 2.2859 for Asian subjects, 0.6741 ± 3.2822 for Caucasian subjects, and 0.9669 ± 2.2268 for African American subjects). However, a sensor that did not prevent crosstalk showed a large error in dark skin subjects (0.8258 ± 2.1603 for Asian subjects, 0.8733 ± 1.9716 for Caucasian subjects, and −3.0591 ± 3.9925 for African Americans). Based on these results, we reiterate the importance of sensor design in the development of pulse oximeters using reflectance-type sensors.

This paper introduces a noise-robust HR estimation algorithm using wrist-type PPG signals that consist of preprocessing block, motion artifact reduction block, and frequency tracking block. The proposed algorithm has not only robustness for motion noise but also low computational complexity. The proposed algorithm was tested on a data set of 12 subjects and recorded during treadmill exercise in order to verify and compare with other existing algorithms.