LAUSR

A new concept of human-machine interface to control hand prostheses based on magnetic tracking, the myokinetic control interface, has been recently proposed. Control signals are the retrieved displacements of multiple magnets implanted in the limb residual muscles following contraction. In previous works, magnets localization has been achieved following an optimization procedure to find an approximate solution to an analytical model. To simplify and speed up the localization problem, here we employ a data-driven strategy to create mathematical models, which can translate measured magnetic information to desired commands for active prosthetic devices. We employed machine learning models, namely linear and radial basis functions artificial neural networks, due to their inherently parallel architecture. They were developed offline and then implemented on field-programmable gate arrays using customized floating-point operators. We optimized computational precision, execution time, hardware, and energy consumption, as they are essential features in the context of wearable devices. When used to track a single magnet in an anatomical mockup of the human forearm, the proposed data-driven strategy achieved a tracking accuracy of 720m 95% of the time and latency of 12.07s. In addition, the proposed system architecture is expected to require a low energy consumption compared to previous solutions. The outcomes of this work encourage further research on improving the devised methods to deal with multiple magnets simultaneously.