Print Email Facebook Twitter Reinforcement Learning for Flight Control Title Reinforcement Learning for Flight Control: Evaluating Handling Qualities and Stability Properties of the PH-LAB Author Jansen, Hidde (TU Delft Aerospace Engineering) Contributor van Kampen, E. (mentor) Pavel, M.D. (graduation committee) Mooij, E. (graduation committee) Konatala, R. (graduation committee) Degree granting institution Delft University of Technology Programme Aerospace Engineering | Control & Simulation Date 2024-04-12 Abstract Reinforcement Learning applied to flight control has shown to have several benefits over classical, linear flight controllers, as it eliminates the need for gain scheduling and it could provide fault-tolerance. The application to civil aviation in practice, however, is non-existent as there are multiple safety concerns. This research demonstrates the evaluation of longitudinal Handling Qualities of the Soft Actor-Critic Deep Reinforcement Learning framework with the aim to translate the unpredictable black box of Reinforcement Learning into classical flight control terminology. The framework is applied to a pitch rate command system of a jet aircraft and shows robustness to off-nominal flight conditions, center of gravity shifts and biased sensor noise. Accurate tracking performance is achieved, while adhering to Level 1 longitudinal Handling Qualities for all conditions. Subject Deep Reinforcement LearningIntelligent Flight ControlHandling QualitiesFlight Control SystemsMachine Learning To reference this document use: http://resolver.tudelft.nl/uuid:13a77d65-116f-44ca-aef8-6978e2185640 Part of collection Student theses Document type master thesis Rights © 2024 Hidde Jansen Files PDF Thesis_Hidde_Jansen_Final.pdf 8.28 MB Close viewer /islandora/object/uuid:13a77d65-116f-44ca-aef8-6978e2185640/datastream/OBJ/view