Activity recognition of the torso based on surface electromyography for exoskeleton control

Abstract This paper presents an activity mode recognition approach to identify the motions of the human torso. The intent recognizer is based on decision tree classification in order to leverage its computational efficiency. The recognizer uses surface electromyography as the input and CART (classification and regression tree) as the classifier. The experimental results indicate that the recognizer can extract the user's intent within 215 ms, which is below the threshold a user will perceive. The approach achieves a low recognition error rate and a user-unperceived latency by using sliding overlapped analysis window. The intent recognizer is envisioned to a part a high-level supervisory controller for a powered backbone exoskeleton.

[1]  R.N. Scott,et al.  A new strategy for multifunction myoelectric control , 1993, IEEE Transactions on Biomedical Engineering.

[2]  Elisabeth André,et al.  EMG-based hand gesture recognition for realtime biosignal interfacing , 2008, IUI '08.

[3]  Adrian D. C. Chan,et al.  A Gaussian mixture model based classification scheme for myoelectric control of powered upper limb prostheses , 2005, IEEE Transactions on Biomedical Engineering.

[4]  Mohammad Hassan Moradi,et al.  Evaluation of the forearm EMG signal features for the control of a prosthetic hand. , 2003, Physiological measurement.

[5]  Guanglin Li,et al.  Principal Components Analysis Preprocessing for Improved Classification Accuracies in Pattern-Recognition-Based Myoelectric Control , 2009, IEEE Transactions on Biomedical Engineering.

[6]  Kevin B. Englehart,et al.  A robust, real-time control scheme for multifunction myoelectric control , 2003, IEEE Transactions on Biomedical Engineering.

[7]  Hassan Ghasemzadeh,et al.  A Body Sensor Network With Electromyogram and Inertial Sensors: Multimodal Interpretation of Muscular Activities , 2010, IEEE Transactions on Information Technology in Biomedicine.

[8]  S. Dosen,et al.  Rule-based control of walking by using decision trees and practical sensors , 2008, 2008 9th Symposium on Neural Network Applications in Electrical Engineering.

[9]  He Huang,et al.  A Strategy for Identifying Locomotion Modes Using Surface Electromyography , 2009, IEEE Transactions on Biomedical Engineering.

[10]  G.F. Inbar,et al.  Classification of finger activation for use in a robotic prosthesis arm , 2002, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[11]  Michael Goldfarb,et al.  Multiclass Real-Time Intent Recognition of a Powered Lower Limb Prosthesis , 2010, IEEE Transactions on Biomedical Engineering.

[12]  Blair A. Lock,et al.  Determining the Optimal Window Length for Pattern Recognition-Based Myoelectric Control: Balancing the Competing Effects of Classification Error and Controller Delay , 2011, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[13]  Michael Goldfarb,et al.  Volitional Control of a Prosthetic Knee Using Surface Electromyography , 2011, IEEE Transactions on Biomedical Engineering.

[14]  Kongqiao Wang,et al.  Automatic recognition of sign language subwords based on portable accelerometer and EMG sensors , 2010, ICMI-MLMI '10.

[15]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[16]  Todd R Farrell,et al.  Determining delay created by multifunctional prosthesis controllers. , 2011, Journal of rehabilitation research and development.

[17]  Gary Kamen,et al.  Essentials of Electromyography , 2009 .