Optimal mapping of torus self‐organizing map for human forearm motion discrimination on the basis of myoelectric signals

Abstract

This paper describes an optimal mapping of the torus self‐organizing map for human forearm motion discrimination on the basis of myoelectric signals. This study uses the torus self‐organizing map (torus SOM) for motion discrimination. The normal SOM identifies input data of the same feature group by using all units of the map. But there is then a possibility of misrecognition of motion around the boundary lines of the motion groups. Therefore, this study proposes a mapping method of SOM in which learning units of the same motion concentrate on one local range and the learning unit groups of each motion are sufficiently separated. As a result, the variance in the same motion group becomes small and the variance between motion groups becomes large. Experiments on a myoelectric hand simulator show the effectiveness of the proposed motion discrimination method. © 2012 Wiley Periodicals, Inc. Electron Comm Jpn, 95(6): 24–31, 2012; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ecj.11369

The role of muscle synergies in myoelectric control: trends and challenges for simultaneous multifunction control

“…(f) Nonlinear projections. Self organizing feature maps create nonlinear projections of EMG features to obtain task-specific synergies [112,182,183]. Despite enhanced separation of classes, nonlinear projections are subject to overfitting the data, which increases sensitivity to transient changes in sEMG signals.…”

A hybrid BMI-based exoskeleton for paresis: EMG control for assisting arm movements

“…(f) Nonlinear projections. Self organizing feature maps create nonlinear projections of EMG features to obtain task-specific synergies [112,182,183]. Despite enhanced separation of classes, nonlinear projections are subject to overfitting the data, which increases sensitivity to transient changes in sEMG signals.…”