Sensor fusion-based UAV localization system for low-cost autonomous inspection in extremely confined environments

Yang, Beiya and Yang, Erfu and Yu, Leijian and Niu, Cong (2024) Sensor fusion-based UAV localization system for low-cost autonomous inspection in extremely confined environments. IEEE Sensors Journal, 24 (14). pp. 22144-22155. ISSN 1530-437X (https://doi.org/10.1109/jsen.2024.3355103)

[thumbnail of Sensor fusion based UAV localisation system for low cost autonomous inspection in extremely confined environments]
Preview
 Text. Filename: Sensor_fusion_based_UAV_localisation_system_for_low_cost_autonomous_inspection_in_extremely_confined_environments.pdf
Accepted Author Manuscript
License: Creative Commons Attribution 4.0 logo
Download (34MB)| Preview

Abstract

Utilizing the unmanned aerial vehicle (UAV) for autonomous inspection in extremely confined environments has become a much sought research area, due to the pressing industrial needs. To provide the high-accuracy position information of UAV in such environments, the ultrawideband (UWB)-based localization technology has been one of the ideal candidates. However, the unpredictable propagation condition and the geomagnetic disturbances for inertial measurement unit (IMU) will all have huge impact on the positioning performance with the single UWB and the IMU/UWB-based loosely coupled (LC) sensor fusion approach. Therefore, a tightly coupled adaptive extended Kalman filter (TC-AEKF)-based sensor fusion UAV localization system is proposed and developed in this article for autonomous inspection in extremely confined environments. The proposed system can attain high-accuracy localization of UAV with 0.097-m median error, 0.167-m 95th percentile error, and 0.039-m average standard deviation (STD). The drift led by the geomagnetic disturbances for IMU is greatly reduced with the estimation error of the roll, pitch, and yaw angle to be 2.15°, 1.54°, and 4.58°, respectively. Finally, an autonomous inspection experiment has been conducted to prove the effectiveness of the proposed system and algorithm for actual application, and the video from it has been attached in https://youtu.be/nEgvwbIVRRk .

CORE (COnnecting REpositories)