0
Journal of Biomechanical Engineering | Volume 138 | Issue 9 | research-article
Research Papers

Human Arm Motion Tracking by Orientation-Based Fusion of Inertial Sensors and Kinect Using Unscented Kalman Filter

[+] Author and Article Information
Arash Atrsaei

Department of Mechanical Engineering,
Sharif University of Technology,
Tehran 1458889694, Iran
e-mail: atrsaei@mech.sharif.edu

Hassan Salarieh

Department of Mechanical Engineering,
Sharif University of Technology,
Tehran 1458889694, Iran
e-mail: salarieh@sharif.edu

Aria Alasty

Department of Mechanical Engineering,
Sharif University of Technology,
Tehran 1458889694, Iran
e-mail: aalasti@sharif.edu

1Corresponding author.

Manuscript received March 15, 2016; final manuscript received July 2, 2016; published online August 2, 2016. Assoc. Editor: Pasquale Vena.

J Biomech Eng 138(9), 091005 (Aug 02, 2016) (13 pages) Paper No: BIO-16-1097; doi: 10.1115/1.4034170 History: Received March 15, 2016; Revised July 02, 2016
Article
References
Figures
Tables

Due to various applications of human motion capture techniques, developing low-cost methods that would be applicable in nonlaboratory environments is under consideration. MEMS inertial sensors and Kinect are two low-cost devices that can be utilized in home-based motion capture systems, e.g., home-based rehabilitation. In this work, an unscented Kalman filter approach was developed based on the complementary properties of Kinect and the inertial sensors to fuse the orientation data of these two devices for human arm motion tracking during both stationary shoulder joint position and human body movement. A new measurement model of the fusion algorithm was obtained that can compensate for the inertial sensors drift problem in high dynamic motions and also joints occlusion in Kinect. The efficiency of the proposed algorithm was evaluated by an optical motion tracker system. The errors were reduced by almost 50% compared to cases when either inertial sensor or Kinect measurements were utilized.
Copyright © 2016 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

The proposed human model (a) human body model and (b) human arm model

+The proposed human model (a) human body model and (b) human arm model
View Large  |  Save Figure  |  Download Slide (.ppt)
The proposed human model (a) human body model and (b) human arm model
Grahic Jump Location
Fig. 3

The flowchart diagram of the proposed algorithm

+The flowchart diagram of the proposed algorithm
View Large  |  Save Figure  |  Download Slide (.ppt)
The flowchart diagram of the proposed algorithm
Grahic Jump Location
Fig. 2

The angle between Kinect x axis and the line passing through the shoulder joints

+The angle between Kinect x axis and the line passing through the shoulder joints
View Large  |  Save Figure  |  Download Slide (.ppt)
The angle between Kinect x axis and the line passing through the shoulder joints
Grahic Jump Location
Fig. 4

The sensors setup

+The sensors setup
View Large  |  Save Figure  |  Download Slide (.ppt)
The sensors setup
Grahic Jump Location
Fig. 12

The effect of utilizing the second measurement equation (a) without constraint and (b) with constraint

+The effect of utilizing the second measurement equation (a) without constraint and (b) with constraint
View Large  |  Save Figure  |  Download Slide (.ppt)
The effect of utilizing the second measurement equation (a) without constraint and (b) with constraint
Grahic Jump Location
Fig. 5

The sampling frequency of the inertial sensor (long lines), which is constant and Kinect (short lines), which is varying

+The sampling frequency of the inertial sensor (long lines), which is constant and Kinect (short lines), which is varying
View Large  |  Save Figure  |  Download Slide (.ppt)
The sampling frequency of the inertial sensor (long lines), which is constant and Kinect (short lines), which is varying
Grahic Jump Location
Fig. 6

The fusion results for the first stage of the test (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z

+The fusion results for the first stage of the test (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
View Large  |  Save Figure  |  Download Slide (.ppt)
The fusion results for the first stage of the test (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
Grahic Jump Location
Fig. 7

The inertial sensors results for the first stage of the test without data fusion (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z

+The inertial sensors results for the first stage of the test without data fusion (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
View Large  |  Save Figure  |  Download Slide (.ppt)
The inertial sensors results for the first stage of the test without data fusion (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
Grahic Jump Location
Fig. 8

Kinect results for the first stage of the test without data fusion, (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z

+Kinect results for the first stage of the test without data fusion, (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
View Large  |  Save Figure  |  Download Slide (.ppt)
Kinect results for the first stage of the test without data fusion, (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
Grahic Jump Location
Fig. 9

The quaternion rotation of the forearm by the fusion method

+The quaternion rotation of the forearm by the fusion method
View Large  |  Save Figure  |  Download Slide (.ppt)
The quaternion rotation of the forearm by the fusion method
Grahic Jump Location
Fig. 10

The quaternion rotation of the upper arm by the fusion method

+The quaternion rotation of the upper arm by the fusion method
View Large  |  Save Figure  |  Download Slide (.ppt)
The quaternion rotation of the upper arm by the fusion method
Grahic Jump Location
Fig. 11

The norm of the accelerometers

+The norm of the accelerometers
View Large  |  Save Figure  |  Download Slide (.ppt)
The norm of the accelerometers
Grahic Jump Location
Fig. 13

The second stage of the test

+The second stage of the test
View Large  |  Save Figure  |  Download Slide (.ppt)
The second stage of the test
Grahic Jump Location
Fig. 14

Kinect results for the second stage of the test without data fusion (a) shoulder X, (b) shoulder Y, (c) shoulder Z, (d) elbow X, (e) elbow Y, (f) elbow Z, (g) wrist X, (h) wrist Y, and (i) wrist Z

+Kinect results for the second stage of the test without data fusion (a) shoulder X, (b) shoulder Y, (c) shoulder Z, (d) elbow X, (e) elbow Y, (f) elbow Z, (g) wrist X, (h) wrist Y, and (i) wrist Z
View Large  |  Save Figure  |  Download Slide (.ppt)
Kinect results for the second stage of the test without data fusion (a) shoulder X, (b) shoulder Y, (c) shoulder Z, (d) elbow X, (e) elbow Y, (f) elbow Z, (g) wrist X, (h) wrist Y, and (i) wrist Z
Grahic Jump Location
Fig. 15

The fusion results for the second stage of the test (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z

+The fusion results for the second stage of the test (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z
View Large  |  Save Figure  |  Download Slide (.ppt)
The fusion results for the second stage of the test (a) elbow X, (b) wrist X, (c) elbow Y, (d) wrist Y, (e) elbow Z, and (f) wrist Z

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Real-Time Prediction Using Kernel Methods and Data Assimilation
Intelligent Engineering Systems through Artificial Neural Networks
Machine Learning Methods for Data Assimilation
Intelligent Engineering Systems through Artificial Neural Networks, Volume 20
Comparing Probabilistic Graphical Model Based and Gaussian Process Based Selections for Predicting the Temporal Observations
Intelligent Engineering Systems through Artificial Neural Networks, Volume 20
The Adviceptron: Giving Advice to the Perceptron
Intelligent Engineering Systems through Artificial Neural Networks, Volume 20
Mitigation of Correlation and Heterogeneity Effects in Hyperspectral Data
Intelligent Engineering Systems through Artificial Neural Networks, Volume 20
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In