Title: Enhanced UAV Navigation in GNSS Denied Environment Using Repeated Dynamics Pattern Recognition
Author(s): S. Zahran, A. Moussa, N. El-Sheimy
Published in: Proceedings of IEEE/ION PLANS 2018
April 23 - 26, 2018
Hyatt Regency Hotel
Monterey, CA
Pages: 1135 - 1142
Cite this article: Zahran, S., Moussa, A., El-Sheimy, N., "Enhanced UAV Navigation in GNSS Denied Environment Using Repeated Dynamics Pattern Recognition," Proceedings of IEEE/ION PLANS 2018, Monterey, CA, April 2018, pp. 1135-1142.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: This paper presents an innovative approach to enhance the navigation of UAVs in GNSS denied environments. Considering the limited space, power, and size of small UAVs, the proposed approach does not require any sensors on the UAV. The typical repeated dynamic patterns of such UAVs are related to the actuators to offer a useful information for estimating the UAV navigation states. Machine learning (ML) classifier has been employed to detect these repeated dynamic patterns, then according to the detected pattern, an appropriate constraint/update is utilized to enhance the navigation solution through EKF to obtain a better estimate of the UAV states. Different test scenarios where conducted to verify the ability of the proposed approach to aid the INS solution during GNSS signal outages. The solution after fusing the Vehicle Model (VM) is enhanced by 98% compared to low cost stand-alone IMU solution.