No abstract

No abstract

A new method for adaptive-fuzzy control achieves knowledge of the model or modeling error (respectively). stabilization of a quadrotor helicopter in the presence of The technique of e-modification does not. However, in the sinusoidal wind disturbance. Techniques traditionally used in quadrotor simulation an adaptive-fuzzy control technique that adaptive control for robust parameter updates may not be q q sufficient for fuzzy schemes. In particular, e-modification may relies on e-modification for robustness may not perform result in the fuzzy-membership centers drifting to large values satisfactorily. Performance must be sacrificed to halt the when persistent oscillations are present in the input. These center drift. large values can cause control signal chatter, which can be We propose a new robust method that does not require undesirable or even cause instability if they excite unmodeled a sacrifice of performance to halt the center drift. An aldynamics. A new method for robust updates is proposed that as e of performanctoh center drt. Anial prevents this drift in the fuzzy membership centers. In the new ternate set of membersh functon centers i used to guide method, a set of alternate membership function centers guides the adaptation. The alternate set is trained to approximate the adaptation process in order to prevent drift. A Lyapunovsame output without exhibiting any center drift. The fuzzy stability proof ensures the uniform ultimate boundedness of all decoding membership function centers used in the control are signals. A simulation of a quadrotor helicopter demonstrates then kept close to these alternate centers in order to prevent the high performance and robust stability of the new method. drift while maintaining a high level of performance.

No abstract

No abstract