Neural network observer based LPV fault tolerant control of a flying-wing aircraft

dc.contributor.authorLiu, Shiqian
dc.contributor.authorLyv, Weizhi
dc.contributor.authorYang, Congjie
dc.contributor.authorWang, Feiyue
dc.contributor.authorWhidborne, James F.
dc.date.accessioned2024-01-11T15:38:27Z
dc.date.available2024-01-11T15:38:27Z
dc.date.issued2023-11-30
dc.description.abstractFor the problem of fault tolerant trajectory tracking control for a large Flying-Wing (FW) aircraft with Linear Parameter-Varying (LPV) model, a gain scheduled H ∞ controller is designed by dynamic output feedback. Robust synthesis of this gain scheduled H ∞ control is carried out by an affine Parameter Dependent Lyapunov Function (PDLF). The problem of trajectory tracking control for the LPV plant is transformed into solving an infinite number of linear matrix inequalities by the PDLF design, and the linear matrix inequalities are solved by convex optimization techniques. To overcome model uncertainties due to linearization and external disturbances, a radial basis function neural network disturbance observer is proposed, and to estimate actuator faults, an LPV fault estimator is designed. Furthermore, a composite controller is proposed to realize fault tolerant trajectory tracking control, which combines the LPV control with the fault estimator and disturbance observer, as well as an active-set based control allocation to avoiding actuator saturation. The approach is tested by simulation of two scenarios that show responses of the altitude, speed and heading angle to (i) unknown disturbances and (ii) actuator faults. The results show that the proposed neural network observer based LPV control has better performances for both disturbance rejecting and fault-tolerant trajectory tracking.en_UK
dc.identifier.citationLiu S, Lyv W, Yang C, et al., (2023) Neural network observer based LPV fault tolerant control of a flying-wing aircraft. Available online 30 November 2023en_UK
dc.identifier.issn2379-8858
dc.identifier.urihttps://doi.org/10.1109/TIV.2023.3337863
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/20644
dc.language.isoenen_UK
dc.publisherIEEEen_UK
dc.rightsAttribution-NonCommercial 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/*
dc.subjectFlying-wing aircraften_UK
dc.subjectlinear parameter-varying systemen_UK
dc.subjectneural networken_UK
dc.subjectdisturbance observeren_UK
dc.subjectfault tolerant controlen_UK
dc.titleNeural network observer based LPV fault tolerant control of a flying-wing aircraften_UK
dc.typeArticleen_UK

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
LPV_fault_tolerant_control-2023.pdf
Size:
1.77 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Item-specific license agreed upon to submission
Description: