The potential of electromyography to aid personal navigation

This paper reports on research to explore the potential for using electromyography (EMG) measurements in pedestrian navigation. The aim is to investigate whether the relationship between human motion and the activity of skeletal muscles in the leg might be used to aid other positioning sensors, or even to determine independently the path taken by a pedestrian. The paper describes an exercise to collect sample EMG data alongside leg motion data, and the subsequent analysis of this data set using machine learning techniques to infer motion from a set of EMG sensors. The sample data set included measurements from multiple EMG sensors, a camera-based motion tracking system and a foot mounted inertial sensor. The camera based motion tracking system at MMU allowed many targets on the subjects lower body to be tracked in a small (3m x 3m x 3m) volume to millimetre accuracy. Processing the data revealed a strong, but not trivial, relation-ship between leg muscle activity and motion. Each type of motion involves many different muscles, and it is not possible to conclude merely from the triggering of any single muscle that a particular type of motion has occurred. For instance, a similar set of leg muscles is involved in both forward and backward steps. It is the precise sequencing, duration and magnitude of multiple muscle activity that allows us to determine what type of motion has occurred. Preliminary analyses of the data suggest that subsets of the EMG sensors can be used to distinguish, for instance, forward motion from backward motion, and it is expected that further analysis will reveal additional correlations that will be useful in inferring the subjects motion in more detail. This paper will introduce the EMG personal navigation con-cept, describe the data collected, explore the machine learning techniques applied to the dataset, and present the results of the analysis.

[1]  T. Moore,et al.  The use of high sensitivity GPS for initialisation of a foot mounted inertial navigation system , 2012, Proceedings of the 2012 IEEE/ION Position, Location and Navigation Symposium.

[2]  Eric Foxlin,et al.  Pedestrian tracking with shoe-mounted inertial sensors , 2005, IEEE Computer Graphics and Applications.

[3]  Yuwei Chen,et al.  Comparison of EMG-based and Accelerometer-based Speed Estimation Methods in Pedestrian Dead Reckoning , 2011, Journal of Navigation.

[4]  Ashutosh Kumar Singh,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2010 .

[5]  H. Devries.  MUSCLES ALIVE-THEIR FUNCTIONS REVEALED BY ELECTROMYOGRAPHY , 1976 .

[6]  Hui Ding,et al.  Querying and mining of time series data: experimental comparison of representations and distance measures , 2008, Proc. VLDB Endow..

[7]  Yuwei Chen,et al.  Sensing strides using EMG signal for pedestrian navigation , 2011 .

[8]  Jaap H van Dieën,et al.  Methodological aspects of SEMG recordings for force estimation--a tutorial and review. , 2010, Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology.

[9]  E L Morin,et al.  Sampling, noise-reduction and amplitude estimation issues in surface electromyography. , 2002, Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology.

[10]  Donald J. Berndt,et al.  Using Dynamic Time Warping to Find Patterns in Time Series , 1994, KDD Workshop.

[11]  K. Abdulrahim,et al.  Using Constraints for Shoe Mounted Indoor Pedestrian Navigation , 2011, Journal of Navigation.

[12]  Patrick Robertson,et al.  Simultaneous localization and mapping for pedestrians using only foot-mounted inertial sensors , 2009, UbiComp.

[13]  Meinard Müller,et al.  Information retrieval for music and motion , 2007 .

[14]  M. Dalakas,et al.  Review: An update on inflammatory and autoimmune myopathies , 2011, Neuropathology and applied neurobiology.

[15]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[16]  Pascal Madeleine,et al.  Advanced biofeedback from surface electromyography signals using fuzzy system , 2010, Medical & Biological Engineering & Computing.

[17]  Dario Farina,et al.  Myoelectric control in neurorehabilitation. , 2010, Critical reviews in biomedical engineering.

[18]  S. Gandevia,et al.  The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. , 2012, Physiological reviews.