A Height-Measurement Algorithm for Rotor UAVs Based on Multilayer Space-Time Fusion

• Published in: IEEE sensors journal Vol. 24; no. 8; pp. 12642 - 12653
• Main Authors: Huang, He, Xie, Feiyu, Niu, Mingbo, Miah, Md. Sipon, Wang, Huifeng, Zhang, Panpan, Wang, Jun
• Format: Journal Article
• Language: English
• Published: New York IEEE 15.04.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerometers; Accuracy; Algorithms; Autonomous aerial vehicles; Data integration; Error reduction; Flight; Height measurement; Low altitude; multilayer; Multilayers; Multisensor fusion; Parameter identification; Rotors; Sea measurements; Sensor fusion; sensor information fusion; Sensors; Spacetime; Stability; Time measurement; unmanned aerial vehicle (UAV); Unmanned aerial vehicles
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2024.3368198

Unmanned aerial vehicle (UAV) is widely used to assist a variety of low-altitude monitoring, sensing, and data relay tasks. Height measurement technology is the primary premise for UAV to complete tasks. Existing UAVs use a single sensor to measure the flight height. Current one-step multisensor fusion method is simple but can be inaccurate in practice while providing limited improvement on measurement accuracy. A dual layer information fusion height measurement algorithm is designed to combat these problems. In the first fusion layer, with a small amount of data used in UAV measurements, an adaptive two-step delayed space-time algorithm is proposed. This algorithm utilizes heterogeneous historical data to improve measurement accuracy. The second fusion layer integrates adaptive explicit complementary filters based on parameter identification by using accelerometer data. The obtained accurate height measurements of the rotor UAV during flight and hover ensures rotor UAV stability. Compared with the one-step delay fusion method, the root mean square error of the proposed algorithm is found to be reduced by 62.4% and the maximum error is also reduced by 85.4%. Experimental results demonstrated that the proposed algorithm is with low complexity and easy to implement with enhanced accuracy and stability.