A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Motion Control of Drives for Prosthetic Hand Using Continuous Myoelectric Signals
Abstract In this paper the authors present motion control of a prosthetic hand, through continuous myoelectric signal acquisition, classification and actuation of the prosthetic drive. A four channel continuous electromyogram (EMG) signal also known as myoelectric signals (MES) are acquired from the abled-body to classify the six unique movements of hand and wrist, viz, hand open (HO), hand close (HC), wrist flexion (WF), wrist extension (WE), ulnar deviation (UD) and radial deviation (RD). The classification technique involves in extracting the features/pattern through statistical time domain (TD) parameter/autoregressive coefficients (AR), which are reduced using principal component analysis (PCA). The reduced statistical TD features and or AR coefficients are used to classify the signal patterns through k nearest neighbour (kNN) as well as neural network (NN) classifier and the performance of the classifiers are compared. Performance comparison of the above two classifiers clearly shows that kNN classifier in identifying the hidden intended motion in the myoelectric signals is better than that of NN classifier. Once the classifier identifies the intended motion, the signal is amplified to actuate the three low power DC motor to perform the above mentioned movements.
Motion Control of Drives for Prosthetic Hand Using Continuous Myoelectric Signals
Abstract In this paper the authors present motion control of a prosthetic hand, through continuous myoelectric signal acquisition, classification and actuation of the prosthetic drive. A four channel continuous electromyogram (EMG) signal also known as myoelectric signals (MES) are acquired from the abled-body to classify the six unique movements of hand and wrist, viz, hand open (HO), hand close (HC), wrist flexion (WF), wrist extension (WE), ulnar deviation (UD) and radial deviation (RD). The classification technique involves in extracting the features/pattern through statistical time domain (TD) parameter/autoregressive coefficients (AR), which are reduced using principal component analysis (PCA). The reduced statistical TD features and or AR coefficients are used to classify the signal patterns through k nearest neighbour (kNN) as well as neural network (NN) classifier and the performance of the classifiers are compared. Performance comparison of the above two classifiers clearly shows that kNN classifier in identifying the hidden intended motion in the myoelectric signals is better than that of NN classifier. Once the classifier identifies the intended motion, the signal is amplified to actuate the three low power DC motor to perform the above mentioned movements.
Motion Control of Drives for Prosthetic Hand Using Continuous Myoelectric Signals
Purushothaman, Geethanjali (author) / Ray, Kalyan Kumar (author)
2014-12-24
6 pages
Article (Journal)
Electronic Resource
English
Development of Pneumatic Myoelectric Prosthetic Hand with Tendon-Driven Wrist
| British Library Online Contents | 2011
A Quality Function Deployment (QFD) approach to designing a prosthetic myoelectric hand
| DOAJ | 2005
Advances in processing of surface myoelectric signals: Part 1
| Springer Verlag | 1995
Control of a Powered Prosthetic Hand Via a Tracked Glove
| British Library Online Contents | 2015