Abstract
Previous works in the literature have claimed that the characteristics of electromyography (EMG) signals depend on each person, and thus, EMG interfaces need to be carefully calibrated for each user in myoelectric control. In this study, we show that the EMG interface used to estimate the joint torques of a user can be constructed simply by incorporating other users’ data without typical calibration process. To achieve this plug-and-play capability, we introduce the concept of collaborative filtering to estimate the joint torque of a novel user by exploiting the preidentified relationships between motion-body features, including EMG signals, and the joint torques of other users. To validate our proposed approach, we compare the performance of estimating joint torque by the proposed method with that by conventional linear regression models as a baseline. We considered the following two baseline methods. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Linear-own:</i> The parameters of the linear model are calibrated for each subject from his/her own training data. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Linear-others:</i> The parameters of the linear model are calibrated with the other users’ data in which the novel user's data are not included. As a result, the estimated joint torques from our proposed approach reveal a better estimation performance than those from the baseline approaches. Furthermore, we also successfully demonstrate online myoelectric control of an upper limb exoskeleton robot with an attached mannequin arm.
Highlights
M YOELECTRIC interfaces have great potential to provide an intuitive way of controlling external devices for human users
From the aforementioned three properties, we found that the relationship between joint torques and motion-body features can be generalized among different users, where the motion-body features are composed of EMG signals, joint angles, body weight, and limb lengths
In our preliminary study [15], we reported that the collaborative filtering approach is potentially useful for estimating joint torques from EMG signals
Summary
M YOELECTRIC interfaces have great potential to provide an intuitive way of controlling external devices for human users. In the former, such movement classes as hand postures were predicted from electromyographies (EMGs) [3], [4] In the latter, the interfaces were trained to estimate such user control commands as joint torques from EMG signals [5]–[8]. The interfaces were trained to estimate such user control commands as joint torques from EMG signals [5]–[8] For both approaches, calibration procedures are needed to determine the parameters for estimating the user-intended output from EMG signals. These calibrations are usually required for each subject and each experiment, since the relationships between the EMGs and the user-intended control outputs tend to be varied Requiring such a cumbersome procedure inhibits the potential distribution of useful myoelectric interfaces. Minimizing the calibration process is a challenge that must be met for assistive devices [10]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
