Metadata

Close
Metadata

1. Identity statement

Reference Type Thesis or Dissertation (Thesis)

Site mtc-m16.sid.inpe.br

Holder Code isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S

Identifier 6qtX3pFwXQZ3P8SECKy/y9gQA

Repository sid.inpe.br/jeferson/2003/05.14.14.55

Last Update 2019:02.19.19.01.30 (UTC) simone

Metadata Repository sid.inpe.br/jeferson/2003/05.14.14.55.56

Metadata Last Update 2024:12.26.14.39.49 (UTC) administrator

Secondary Key INPE-9878-TDI/874

Citation Key Cardenuto:2003:AeCoCo

Title Aeronaves configuradas por controle do tipo preditivo neural

Alternate Title Aircraft configured by control of the predictive neural type

Course CAP-SPG-INPE-MCT-BR

Year 2003

Secondary Date 20031403

Date 2003-03-14

Access Date 2025, May 10

Thesis Type Tese (Doutorado em Computação Aplicada)

Secondary Type TDI

Number of Pages 144

Number of Files 295

Size 12066 KiB

2. Context

Author Cardenuto, Nei Cardoso

Group CAP-SPG-INPE-MCT-BR

Committee Ramos, Fernando Manuel (presidente) Rios Neto, Atair (orientador) Rosa, Reinaldo Roberto Carrara, Valdemir Curvo, Marcelo Yoneyama, Takashi

University Instituto Nacional de Pesquisas Espaciais (INPE)

City São José dos Campos

History (UTC) 2008-08-21 20:58:53 :: administrator -> jefferson :: 2009-04-27 18:47:25 :: jefferson -> administrator :: 2009-05-02 03:09:23 :: administrator -> marciana :: 2009-05-11 14:56:18 :: marciana -> administrator :: 2009-06-16 17:20:55 :: administrator -> marciana :: 2009-06-18 17:14:09 :: marciana -> administrator :: 2009-06-18 17:44:40 :: administrator -> marciana :: 2009-06-19 14:48:00 :: marciana -> administrator :: 2009-06-23 18:37:32 :: administrator -> banon :: 2009-06-23 18:54:39 :: banon -> administrator :: 2009-07-08 21:18:38 :: administrator -> marciana :: 2012-07-18 12:14:55 :: marciana -> administrator :: 2018-06-05 01:20:25 :: administrator -> marciana :: 2003 2018-06-08 14:34:55 :: marciana -> sergio :: 2003 2019-10-08 18:03:15 :: sergio -> simone :: 2003 2019-10-08 18:03:42 :: simone -> administrator :: 2003 2024-12-26 14:39:49 :: administrator -> :: 2003

3. Content and structure

Is the master or a copy? is the master

Content Stage completed

Transferable 1

Keywords neural controle de aeronave predição controle ótimo controle de inclinação neural aircraft control prediction optimal control pitch atitude control

Abstract Este trabalho desenvolve e testa uma abordagem de controle de aeronave em manobras não típicas conhecido como "control configured vehicle", ou veículo (aeronave) configurado a controle, onde a abordagem proposta é controle preditivo neural. As manobras não típicas normalmente são obtidas desacoplando modos normalmente acoplados em dinâmica de voo de aeronaves. A técnica mais usual é a atribuição de auto-estrutura, que trata de alocar simultaneamente autovalores e autovetores, sendo estes determinados de acordo com uma referência pré-estabelecida para o desacoplamento desejado. Na técnica de auto-estrutura, o sistema é linear e seu modelo é conhecido, limitando sua aplicação. A proposta de se utilizar controle preditivo com redes neurais visa obter ao mesmo desacoplamento sem a necessidade do conhecimento do modelo da planta, treinando a rede neural para modelar os modos dinâmicos da aeronave. O desempenho desta abordagem é demonstrado e os resultados são apresentados comparando-se: três redes treinadas, resultados de controle preditivo convencional e neural, gradiente analítico e numérico para a otimização, e horizontes de predição e controle. ABSTRACT: This work develops and tests a predictive neural control technique applied in the aircraft non-typical maneuvers known as the control configured vehicle (CCV). These maneuvers are performed with flight dynamics mode decoupling. The usual approach is the eigenstucture assignment with eigenvalues and eigenvectors placement, which is restricted to a known and linear systems model. The predictive control with neural nets approaches looks at the mode decoupling without the plant model knowledge by using neural net training in order to emulate the aircraft dynamic. The performance of the adopted approach is demonstrated and the results presented by considering and comparing: three neural nets, predictive control and neural predictive control, analytic and numeric gradient in the optimizing calculations, and prediction and control horizons.

Area COMP

Arrangement urlib.net > BDMCI > Fonds > Produção pgr ATUAIS > CAP > Aeronaves configuradas por...

doc Directory Content access

source Directory Content

publicacao.pdf 19/02/2019 16:02 779.6 KiB

agreement Directory Content

TERMO DE DEPOSITO ASSINADO NEI CARDOSO CARDENUTODOUTORADO.pdf 19/02/2019 16:01 338.6 KiB

4. Conditions of access and use

data URL http://urlib.net/ibi/6qtX3pFwXQZ3P8SECKy/y9gQA

zipped data URL http://urlib.net/zip/6qtX3pFwXQZ3P8SECKy/y9gQA

Language pt

Target File publicacao.pdf

User Group administrator marciana sergio simone

Visibility shown

Rightsholder originalauthor yes

Copy Holder SID/SCD

Read Permission allow from all

Update Permission not transferred

5. Allied materials

Next Higher Units 8JMKD2USNNW34T/4CL6M65 8JMKD3MGPCW/3F2PHGS

Citing Item List sid.inpe.br/bibdigital/2013/10.12.22.16 2

Dissemination NTRSNASA; BNDEPOSITOLEGAL.

Host Collection sid.inpe.br/banon/2003/08.15.17.40

6. Notes

Empty Fields academicdepartment affiliation archivingpolicy archivist callnumber contenttype creatorhistory descriptionlevel doi e-mailaddress electronicmailaddress format isbn issn label lineage mark mirrorrepository nextedition notes number orcid parameterlist parentrepositories previousedition previouslowerunit progress readergroup resumeid schedulinginformation secondarymark session shorttitle sponsor subject tertiarymark tertiarytype url versiontype