Answers are grounded in this paper and cite their sources.
Human activity recognition has been broadly evaluated in the industry due to ergonomics and optimization purposes. This recognition can provide a human evaluation for several purposes such as motion characteristics, fatigue, and performance analysis. There are some challenges in recognizing the indicators that are used to evaluate the characteristics of motion. The automatic system needs to understand what kind of human activity is in the production system. This article uses motion classification to estimate human motion characteristics. It is known that humans dominantly use their hands in the assembly process. To support it, the electromyography (EMG) sensor was used to detect Biceps Brachii and Flexor Digitorum muscle activity. This article uses basic motion classification to estimate human activity. Four motions were classified using a machine learning algorithm. The result informs that the accuracy reached 89.5%. The evaluation using 2-channel EMG provides good performance in classifying the human motions. It indicates that muscle activity could be used as information for the system to recognize human activity status in the assembly system. This evaluation can also reduce the number of sensors that are used in human activity recognition. This will be useful input for human performance evaluation and prediction in the smart manufacturing system.
No takes yet. Share an insight, caveat, or question.
Lukodono et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: