Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure

碩士 === 國立成功大學 === 航空太空工程學系 === 106 === This paper aims to propose a strategy for quad-rotors flight control under rotor failure. The proposed control strategy consists of two stages – FDI (fault detection and isolation) part and the FTC (fault tolerant control) part. The control method takes normal...

Full description

Bibliographic Details
Main Authors: Yu-HsuanLien, 練羽軒
Other Authors: Chao-Chung Peng
Format: Others
Language:en_US
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/r67wva
id ndltd-TW-106NCKU5295035
record_format oai_dc
spelling ndltd-TW-106NCKU52950352019-11-09T05:23:02Z http://ndltd.ncl.edu.tw/handle/r67wva Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure 適應性觀測器於四旋翼致動器失效下之障誤偵測與容錯控制 Yu-HsuanLien 練羽軒 碩士 國立成功大學 航空太空工程學系 106 This paper aims to propose a strategy for quad-rotors flight control under rotor failure. The proposed control strategy consists of two stages – FDI (fault detection and isolation) part and the FTC (fault tolerant control) part. The control method takes normal flight as well as abnormal flight into consideration. The algorithm that switches the control law between the aforementioned two cases relies on the FDI result. As a consequence, the accuracy of the FDI becomes one of the important issues for field safety flight control. To achieve the FDI, a model based method for fault detection and fault estimation is developed. As for the fault detection, an adaptive boundary of the residues generated from the nonlinear observer is applied, and the fault alarm is going to be triggered if any of the residues exceeds prescribed upper/lower bounds. On the other hands, the update law of fault estimation is formed within the stability analysis of state estimation error, which requires the application of the nonlinear observer. With the combination of the results from fault detection and fault estimation, the judgment in deciding whether the fault actually happens or not could be made. While fault happening, a control strategy that stabilizes the system and keeps the trajectory ability of the quad-rotor is required. To this aim, a coordinate transformation scheme is applied, which corrects the target angles for the quad-rotor to follow a desired trajectory. Furthermore, in order to achieve robust flight during the external wind perturbation, the sliding mode control (SMC) theory is used. Finally, simulations are applied to illustrate the effectiveness and feasibility of the fault control strategy. Numerical results show that the proposed method is able to achieve robust flight even in the presence of rotor failures. Chao-Chung Peng 彭兆仲 2018 學位論文 ; thesis 97 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立成功大學 === 航空太空工程學系 === 106 === This paper aims to propose a strategy for quad-rotors flight control under rotor failure. The proposed control strategy consists of two stages – FDI (fault detection and isolation) part and the FTC (fault tolerant control) part. The control method takes normal flight as well as abnormal flight into consideration. The algorithm that switches the control law between the aforementioned two cases relies on the FDI result. As a consequence, the accuracy of the FDI becomes one of the important issues for field safety flight control. To achieve the FDI, a model based method for fault detection and fault estimation is developed. As for the fault detection, an adaptive boundary of the residues generated from the nonlinear observer is applied, and the fault alarm is going to be triggered if any of the residues exceeds prescribed upper/lower bounds. On the other hands, the update law of fault estimation is formed within the stability analysis of state estimation error, which requires the application of the nonlinear observer. With the combination of the results from fault detection and fault estimation, the judgment in deciding whether the fault actually happens or not could be made. While fault happening, a control strategy that stabilizes the system and keeps the trajectory ability of the quad-rotor is required. To this aim, a coordinate transformation scheme is applied, which corrects the target angles for the quad-rotor to follow a desired trajectory. Furthermore, in order to achieve robust flight during the external wind perturbation, the sliding mode control (SMC) theory is used. Finally, simulations are applied to illustrate the effectiveness and feasibility of the fault control strategy. Numerical results show that the proposed method is able to achieve robust flight even in the presence of rotor failures.
author2 Chao-Chung Peng
author_facet Chao-Chung Peng
Yu-HsuanLien
練羽軒
author Yu-HsuanLien
練羽軒
spellingShingle Yu-HsuanLien
練羽軒
Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure
author_sort Yu-HsuanLien
title Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure
title_short Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure
title_full Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure
title_fullStr Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure
title_full_unstemmed Adaptive Observer Based Fault Detection and Fault-Tolerant Control of Quadrotors under Rotor Failure
title_sort adaptive observer based fault detection and fault-tolerant control of quadrotors under rotor failure
publishDate 2018
url http://ndltd.ncl.edu.tw/handle/r67wva
work_keys_str_mv AT yuhsuanlien adaptiveobserverbasedfaultdetectionandfaulttolerantcontrolofquadrotorsunderrotorfailure
AT liànyǔxuān adaptiveobserverbasedfaultdetectionandfaulttolerantcontrolofquadrotorsunderrotorfailure
AT yuhsuanlien shìyīngxìngguāncèqìyúsìxuányìzhìdòngqìshīxiàoxiàzhīzhàngwùzhēncèyǔróngcuòkòngzhì
AT liànyǔxuān shìyīngxìngguāncèqìyúsìxuányìzhìdòngqìshīxiàoxiàzhīzhàngwùzhēncèyǔróngcuòkòngzhì
_version_ 1719288853962948608