Fast algorithms for elbow motion intention detection by using Levant’s differentiation of biceps and triceps EMG signals

Electromyography (EMG) signals are widely used for predicting human movement intention in robotic assistive devices that improve the quality of people’s lives with motor problems. One of the current challenges controlling such devices is achieving a natural interaction between the device and the user. Among the different techniques proposed in the literature for this purpose, time-domain feature extraction methods have been widely used to process EMG signals, detect movement intention, and thus command the orthosis operation. However, the most common algorithms applied in motion detection introduce additional delays, which may cause the movement of the orthosis to feel unnatural from the user’s perspective. Therefore, in this work, a proposal is made to use robust differentiator algorithms to extract features from EMG signals that allow fast detection of movement intention. Additionally, the use of Levant’s robust differentiators and an adaptive network-based fuzzy inference system (ANFIS) model is proposed to estimate the upper-limb torque exerted by a user as a continuous signal without introducing additional delays. Experimental results show that robust differentiator algorithms can significantly reduce the latency between the detection movement intention and the real movement without compromising accuracy.

https://orcid.org/0000-0003-4191-4143