Load recognition based on sEMG of human upper limbs

This article mainly studies the human upper limb forearm, upper arm, upper arm and forearm, and left and right forearms when snatching a load of 10kg, 20kg, 30kg, 40kg, 50kg, and 60kg respectively. By collecting the sEMG signals in these four situations, and from time to time Extract 5 common features of sEMG signal from the domain as the feature vector of the original signal, and then use decision tree, random forest, LDA, K nearest neighbor, SVM to perform pattern recognition on the feature vector, and select the sEMG feature that is more suitable for load identification and recognition algorithm. Finally, the best method among the above four load identification methods is obtained through comparative analysis of experimental data.