МЕТОД ОПТИМИЗАЦИИ КЛАССИФИКАТОРА ЖЕСТОВ

Abstract

The work is devoted to the study of the possibility of optimizing the process of synthesis ofgesture classifiers by selecting the most significant channels of electromyographic (EMG) activityof the muscles of the forearm. The first part of the study is devoted to the development and analysisof the performance of gesture classifiers with a different number of EMG channels, ranked bysignificance based on the Pearson criterion. The solution of the problem of classification of gesturesby EMG signals was first implemented on the basis of ensembles of decision trees trained bythe gradient boosting method. For this, software was developed that allows automatic synthesisand training of gesture classifiers. Next, a series of studies was carried out to find the optimalnumber of EMG channels based on three criteria: the classifier learning rate, the performance ofthe trained model, and the area under the ROC AUC error curve. To do this, a cycle of trainingand testing of the classifier was carried out for data sets recorded at different positions of the electrodeson the forearm. Then, range diagrams of the studied criteria were constructed for variousnumbers of EMG channels involved in the work from 1 to 8, ranked by significance in each of thesamples. It was found that the optimal number of EMG channels involved under the experimentalconditions was 3-6, since a further increase did not lead to a decrease in the classification error,while significantly degrading the performance. The proposed method allows you to automaticallyselect the channels, the electrodes of which are located above the most informative areas of theforearm in case of an accidental change in the position of the sensors. The second part of the workcontains the results of a full-scale experiment to demonstrate the possibility of controlling awheeled robot through EMG analysis.