Multi-Source Adaptive Learning for Fast Control of Prosthetics Hand

We present a benchmark of several existing multi-source adaptive methods on the largest publicly available database of surface electromyography signals for polyarticulated self-powered hand prostheses. By exploiting the information collected over numerous subjects, these methods allow to reduce significantly the training time needed by any new prosthesis user. Our findings provide the biorobotics community with a deeper understanding of adaptive learning solutions for user-machine control and pave the way for further improvements in handprosthetics.