Flight Test
Flight testing encompasses the experimental evaluation of aircraft and unmanned aerial vehicle (UAV) performance, encompassing diverse objectives like autonomous navigation, energy-efficient perching, and payload manipulation. Current research emphasizes the development and validation of advanced control algorithms, including reinforcement learning, model predictive control, and nonlinear dynamic inversion, often coupled with data-driven approaches like Gaussian processes and neural networks for improved robustness and efficiency. These advancements are crucial for enhancing safety, autonomy, and operational capabilities in various applications, from urban air mobility to space exploration, by providing rigorous validation of theoretical models and enabling the development of more reliable and sophisticated systems.
Papers
Neural Predictor for Flight Control with Payload
Ao Jin, Chenhao Li, Qinyi Wang, Ya Liu, Panfeng Huang, Fan Zhang
A Plug-and-Play Fully On-the-Job Real-Time Reinforcement Learning Algorithm for a Direct-Drive Tandem-Wing Experiment Platforms Under Multiple Random Operating Conditions
Zhang Minghao, Song Bifeng, Yang Xiaojun, Wang Liang