Research On Prosthetic Hand Control Technology Based On MYO Armband

Surface electromyography(sEMG)is a bioelectrical signal which generated by body muscle contraction.With the tireless efforts of scholars at home and abroad,sEMG has been widely used in clinical testing,rehabilitation engineering and prosthetic hand control and other fields.At present,sEMG-based prosthetic hand control technology has become a hot research.Compared with traditional sensors,MYO armband has many advantages such as no restrictions on the venue,interactive nature,easy to wear and high cost-effective.It is very suitable for the control of prosthetic hand.Therefore,the purpose of this paper is to study a myoelectric prosthetics control technology based on MYO armband,through the algorithm to achieve hand motion pattern recognition and the grasping force of prediction,and verify the accuracy by combining with the prosthetic limb electromyography control system on PC.The main research work is as follows:(1)Research on hand motion pattern recognition.This experiment used six order Butterworth bandpass filter to preprocess the MYO arms which collect sEMG and extract five TD features.In this paper,PCA and BP neural network are used to classify the hand motion patterns.The experimental results show that the recognition rate of hand motion pantterns can reach 99% when feature sample is mapped to 20 dimensions based on PCA.(2)Research on the technology of grasping force prediction.In this experiment,the absolute mean value(MAV)and root mean square(RMS)are selected as the features,taking the eight grades of grasping force as output,a prediction model of grasping force based on BP neural network is established.The experimental results show that the average recognition rate of the grabbing force can reach 93.83% based on the size of grasping force,and it can meet the basic requirements of prosthetic hand control.(3)The system design of myoelectric control for prosthetic hand.This paper has designed the myoelectric control system on MFC.The system can acquire data,and analyze these data to get the activity intention of the hand motion patterns.The system is also able to real-time predict the force size of human finger grabbing,and drive the prosthetic hand by serial port communication.Finally,this scheme is proved that it is feasible through the application of the system.The myoelectric control system can realize the real-time classification of hand motion pattern and the real-time prediction of the grasping force based on sEMG.The extracted action intention and grasping force can be converted into different control commands,which can provide an effective human-computer interaction model based on bioelectrical signals.The main innovation of this system lies in that the high cost-effective of the MYO armband is used to control prosthetic hand.And we realize the online recognition of hand movement patterns and grasping force of prosthetic hand.Moreover,online recognition rate reached 92%.It is convenient for the system to be installed and utilized,and the system holds strong anti-interference ability and high controllability,may meet the needs of the prosthetic hand control of the disabled commendably.