Abstract
With coexisting-cooperative-cognitive robots gradually appearing in daily life, an instinct and efficient human-robot interaction (HRI) is becoming more and more challenging and necessary. Surface electromyography (sEMG) signals, as one of mainstream manners of the interactions, are employed to predict human intentions. In this paper, to provide natural assistance for standing up and sitting down, sEMG signals acquired from active muscles of one’s lower limb are utilized to predict continuous movements. A temporally smoothed multilayer perceptron (MLP) regression scheme is proposed for continuous knee/ankle angles estimation by multi-channel sEMG signals. After correlation analyses of sEMG signals and movements, a series of linear and nonlinear regression models are trained to decode human intentions from pre-processed sEMG. Furthermore, to remove out local fluctuations of direct mappings, temporally smoothed techniques are further implemented as post-processings. In the experiments of standing up and sitting down, extensive results of ten healthy subjects show that a three-layer MLP with the Savitzky-Golay filter achieves the best performance on the mean squared error (MSE, testing: 59.58) and the $R^{2}$ score ( $R^{2}$ , testing: 0.948). The proposed regression scheme is compared with other methods and is also verified by measurements of a high-precision visual motion capture system.
Highlights
Human limbs movements in the macro scope are controlled by muscle fibers’ voluntary contractions stimulated by bioelectrical potentials in the micro scope. These bioelectrical potentials, which are collected from the surface of the skin, are called surface electromyogram signals [1], [2]
OF PRE-PROCESSED surface electromyogram (sEMG) AND THE FEATURE mean absolute value (MAV) Fig. 5 displays one-channel raw sEMG signals of a subject’s trial and their pre-processed results, which are processed by a sequence of key procedures
To balance the frequency difference between the sEMG acquisition and the motion capture, all twelve-channel sEMG signals are subsampled by a sliding window technique as in (3), and extracted by the MAV feature as in (4)
Summary
Human limbs movements in the macro scope are controlled by muscle fibers’ voluntary contractions stimulated by bioelectrical potentials in the micro scope. Z. Li et al.: Temporally Smoothed MLP Regression Scheme for Continuous Knee/Ankle Angles Estimation provide lower limb assistance, using EMG-based feedback controller [10]. As shown, the context of movements are taken into consideration It is an another motivation of our work: further improving sEMG-based joints estimations by using temporally smoothed techniques. Linear and nonlinear data-driven models are identified and used to predict sEMG-based motions, including linear regression models, Bayesian ridge regression model, support vector regression (SVR) model and multilayer perceptrons (MLP) By comparisons of their MSE (mean squared error) and the R2 scores, three-layer MLP outperforms others and achieves the best results. A temporally smoothed MLP regression scheme is proposed for continuous lower limb estimations by using multi-channel sEMG, mainly for fluctuations problems in direct mappings.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call