UAV flight test and its endurance degradation modelling by considering the energy efficiency and flight stability factors

• Main Authors: Wang, Jinlong, Govind, Siddesh, Hu, Xinting, Feroskhan, Mir
• Other Authors: School of Mechanical and Aerospace Engineering
• Format: Conference or Workshop Item
• Language: English
• Published: 2023
• Subjects: Engineering::Aeronautical engineering::Accidents and air safety; Unmanned Aerial Vehicle; Performance Degradation Test
• Online Access: https://hdl.handle.net/10356/172714
• Institution: Nanyang Technological University

This study presents a novel method for UAV degradation modeling based on real-world flight data, emphasizing the introduced factors of flight energy efficiency and flight operating stability. The approach adopts a dual-modeling framework. Firstly, an LSTNet (Long Short-Term Memory Network) model interprets a vast time series dataset from flight logs, focusing on the individual rotator's instantaneous rotation speeds, to forecast a flight efficiency indicator in a many-to-one manner. This predicted efficiency marker, 'FE_KPCA', when combined with other metadata parameters, aids regression models in the estimation of the UAV's flight endurance for the second modelling objective. The experimental design for this study, which produced over 40 hours of manual flight data, serves as a notable contribution and foundation for our findings.