Gait Classification Using Wavelet Descriptors in Pedestrian Navigation

Yunqian Ma and Joel A. Hesch

Abstract:	The design of pedestrian navigation systems for use in GPS-denied scenarios has received significant attention from research community in recent years. Numerous target applications exist, including localization for groups of firefighters, first responders, or soldiers. In these applications, the safety and efficiency of the entire team relies on the availability of accurate position and orientation (pose) estimates of each team member. One approach is to equip each person with a body-mounted Inertial Measurement Unit (IMU). As the person moves, the linear acceleration and rotational velocity measurements can be integrated to obtain a pose estimate. However, the integration of both sensor noise and unknown bias causes the pose estimates to drift quickly. To mitigate the inertial drift errors, an aiding sensor can be employed such as a camera or laser scanner, which provides exteroceptive information about the environment. The person´s pose can be estimated by fusing the integrated IMU signals with environmental cues, such as the locations of nearby landmarks, in order to improve pose-estimation accuracy. While these aiding sensors are typically considered to be essential for accurate, GPS-denied navigation, they often require additional infrastructure (e.g., radio beacons), which increases the complexity, cost, and power requirements of the personal navigation system. In contrast, we exploit the wealth of information available from human gait motion in order to improve localization accuracy. Specifically, we employ wavelet signatures computed from the raw IMU signals (tri-axial gyroscope and tri-axial accelerometer measurements) in order to classify the current gait of the person (e.g., walking, running, crawling) and utilize stochastic constraints on the person’s motion, available from trained motion models, in order to correct their pose estimates. We have tested our approach both in simulation and experimentally to validate its correctness and accuracy in real-time personal navigation scenarios. The key benefits of our approach are that it is computationally inexpensive, flexible amongst many users, and extensible to a wide variety of gaits. Moreover, we do not require any additional sensors, hence reducing the cost, weight, and power requirements of our system.
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:	1328 - 1336
Cite this article:	Ma, Yunqian, Hesch, Joel A., "Gait Classification Using Wavelet Descriptors in Pedestrian Navigation," Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011), Portland, OR, September 2011, pp. 1328-1336.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In