Step-length Estimation Using Wrist-worn Accelerometer and GPS

Wei-Wen Kao, Ching-Kun Chen, and Jing-Shian Lin

Abstract:	Pedestrian Dead-Reckoning (PDR) is useful to provide continuing positioning capability during GPS outages and has many commercial applications, including personal navigation system, LBS, sports and recreations, etc. In many PDR systems IMU can be mounted on human foot or torso positions and mathematical models for step length estimation using IMU sensor data have been successfully developed. However in some cases, wrist-wearable products (such as sport watch for runner training) are preferred and PDR based on wrist mounted sensor is desirable. In this paper a complete framework for step-length estimation using wrist-worn accelerometer and GPS measurements is presented along with experimental results. The physiological characteristics of human wrist during walking and running motions are studied and mathematical models that relate wrist-worn accelerometer measurements with the corresponding step lengths in different human walking and running motions are derived. The wrist-worn sensor step length estimation using these models achieves accuracy comparable to previous results using foot-mounted or torso-mounted sensors. As parameters of the step-length estimation models vary with different persons and motions, tedious works to determine suitable model parameters may be necessary before one can use the estimation algorithm effectively. To deal with this problem, GPS data is combined with sensor measurements so that model parameter estimation as well as better step-length estimation result can be achieved under good GPS reception conditions using Kalman filtering technique. When GPS outage occurs, the estimated model can provide reliable step-length estimation to enable continuous product operation. Test results are presented to demonstrate accuracy of the step-length estimation under various GPS reception environments, using a commercial GPS sport watch with a built-in 3-axis accelerometer.
Published in:	Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011) September 20 - 23, 2011 Oregon Convention Center, Portland, Oregon Portland, OR
Pages:	3274 - 3280
Cite this article:	Kao, Wei-Wen, Chen, Ching-Kun, Lin, Jing-Shian, "Step-length Estimation Using Wrist-worn Accelerometer and GPS," Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011), Portland, OR, September 2011, pp. 3274-3280.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In