Reinforcement adaptive fuzzy control of wing rock phenomena

Wing rock phenomena is a limit cycle roll oscillation experienced by aircraft with slender delta wings at a high angle of attack. A new reinforcement adaptive fuzzy control scheme is proposed for suppressing or tracking wing rock phenomena. A fuzzy logic system is used to approximate a unknown nonlinear dynamics in which a variable universe technique is applied to modify the premise parameters of FLS for improving its interpolation precision, and the derived adaptive law combining with a reinforcement-learning strategy is applied to tune its consequent parameters. The novelty of the proposed control scheme is mainly in its simplicity, robustness and stability so that it can be applied for online learning and real-time control. Three cases of time-varying wing rock control under disturbances are simulated to confirm the effectiveness and robustness of the proposed scheme.