06.1 - Flight Dynamics and Control (Control & Modelling)

FAULT-TOLERANT CONTROL FOR AIRCRAFT WITH STRUCTURAL DAMAGE USING GAUSSIAN PROCESS REGRESSION

D.-W. Lee¹, L.-S. Kim¹, H.-C. Bang¹; ¹KAIST, South Korea

This paper presents the design of fault-tolerant control using Gaussian Process regression to stabilize anrnaircraft with left-wing damage. The structural damage causes the change of mass, moment of inertia, center ofrngravity, and aerodynamic coefficient. These variation parameters and unmodeled dynamics make the systemrnunstable. Hence, we need the data-driven fault-tolerant control approach to compensate for the uncertaintyrnunder unexpected situations. Unlike the parametric adaptive control approach, Gaussian Process regressionrnis a non-parametric method that does not need prior information about the uncertainty. Therefore, we suggestrnthe fault-tolerant control using Gaussian Process regression to estimate the structural fault uncertainty. Thernnumerical simulation results are compared with sliding mode control to analyze the tracking performance andrnrobustness.

Sorry, there is no file available for this paper.