EMG Controlled Bionic Robotic Arm using Artificial Intelligence and Machine Learning

• Main Authors: Satya Sree, K P N V, Bikku, Thulasi, Mounika, S, Ravinder, N, Kumar, M Lakshmana, Prasad, Chitturi
• Format: Conference Proceeding
• Language: English
• ISSN: 2768-0673
• DOI: 10.1109/I-SMAC52330.2021.9640623

Electromyography is a one of a kind methodology for recording and breaking down the electrical action created by muscles, and a Myo-electric controlled prosthetic appendage is an ostensibly controlled fake appendage which is constrained by the electrical signals in-stinctively delivered by the muscle framework itself. Computerized reasoning and Machine learning is exceptionally incredible in each mechanical field alongside biomedical field. The motivation behind this work is to use the force of Machine learning and Deep learning for foreseeing and perceiving hand motions for prosthetic hand from gathering information of muscle exercises. Albeit this innovation as of now exists in the mechanical world yet those are exorbitant and not accessible in non-industrial nations. Thus, planning a savvy prosthetic hand with the boost precision is the significant concentration and objective of this work. We have additionally utilized an informational collection recorded by MyoWare Muscle Sensor which addresses continuous readings from 8 sensors. We have utilized Deep learning with the informational index for foreseeing the accompanying signals which are hand-open, hand-close, circular grasp, and fine-squeeze. Then, at that point, we utilized a few calculations of Machine Learning which are K-closest Neighbor (KNN), Support Vector Machine (SVM), and furthermore the mix of KNN and SVM both for highlight order on information recorded with the 8-anode surface EMG (sEMG) MyoWare Muscle Sensor. Utilizing the blend of SVM and KNN We have achieved a continuous test ac-curacy of 96.33 percent at characterizing the four tokens of our prosthetic hand. This paper likewise addresses 3D displaying of the automated hand and its control framework utilizing Autodesk 3D's Max programming, EMG MyoWare Muscle Sensor, Machine Learning and Deep Learning.