Loading...
Publication
Técnicas de identificação de sistemas para estimar o modelo dinâmico longitudinal e avaliar parâmetros dos controladores de uma aeronave não-tripulada, equipada com um piloto automático Piccolo II
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 9.88 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
A presente dissertação enquadra-se na missão do Centro de Investigação da Academia da Força Aérea (CIAFA) e pretende complementar a ferramenta desenvolvida anteriormente em para o ajuste dos
ganhos dos controladores longitudinais do piloto automático Piccolo II. Inicialmente, são analisadas as principais referências no âmbito de identificação de sistemas que permitem caraterizar a dinâmica longitudinal
de um Unmanned Aerial Vehicle (UAV).
Em seguida, são analisadas as leis de controlo do movimento longitudinal do Piccolo II, referentes ao controlo da altitude e da True Air Speed da aeronave. Uma vez que as estas não são disponibilizadas
pelo fabricante, interpreta-se a documentação do Piccolo II e estuda-se a dissertação de mestrado, que aborda o mesmo tema.
Adicionalmente, identificam-se os modelos da dinãmica longitudinal, que juntamente com as leis de controlo caraterizam a estrutura de controlo do Piccolo II, dos quais se apresentam as respostas estimadas
pelos modelos computacionais em comparação com as obtidas atravês do simulador de voo do Piccolo II.
Poróm, com base na técnica Root Locus e na ferramenta computacional Optimization-Cased Control System Design, procede-se para o ajuste dos ganhos do Piccolo II referentes ao movimento longitudinal.
Através das simulações realizadas com o software do Piccolo II, após o ajuste dos ganhos da lei de controlo do leme de profundidade, registaram melhorias na ordem dos 50% nos tempos de resposta da aeronave, e
de 66% na oscilação da mesma, comparativamente aos ganhos pré-definidos do Piccolo II. Este resultado evidencia o potencial de contribuição deste trabalho para a melhoria da seguran ca e eficiência do voo das
aeronaves do CIAFA.
This dissertation addresses the mission of the Centre of Investigation of the Air Force Academy (CIAFA) and complements the tool that was previously developed in the reference to tune the gains of the automatic pilot system Piccolo II. In addition, it intends to provide more information about system identification to estimate the computational models which characterize the longitudinal dynamics of a UAV, with the intention of systematizing the process of tuning of the gains of the controllers of the automatic pilot system Piccolo II. First and foremost, the control laws of the longitudinal dynamics of the automatic pilot Piccolo II are studied in relation to the altitude and the True Air Speed control as well as the gains which belong to both the control laws. Since the control laws have not been made public by the manufacturer, the documentation of Piccolo II and the work in are analysed in order to make up for the unavailability of the control laws Secondly, the models of the longitudinal dynamics, which characterize the control structure of the Piccolo II together with the control laws, are identified using system identification techniques. Moreover, the estimated models and the recorded dynamics are compared and contrasted. Finally, the Root Locus and Optimization-Cased Control System Design are implemented, making way for the tuning of the gains of the Piccolo II with regard to longitudinal movement. Significant improvement in the performance of the UAV Arcturus T-15 is also shown.
This dissertation addresses the mission of the Centre of Investigation of the Air Force Academy (CIAFA) and complements the tool that was previously developed in the reference to tune the gains of the automatic pilot system Piccolo II. In addition, it intends to provide more information about system identification to estimate the computational models which characterize the longitudinal dynamics of a UAV, with the intention of systematizing the process of tuning of the gains of the controllers of the automatic pilot system Piccolo II. First and foremost, the control laws of the longitudinal dynamics of the automatic pilot Piccolo II are studied in relation to the altitude and the True Air Speed control as well as the gains which belong to both the control laws. Since the control laws have not been made public by the manufacturer, the documentation of Piccolo II and the work in are analysed in order to make up for the unavailability of the control laws Secondly, the models of the longitudinal dynamics, which characterize the control structure of the Piccolo II together with the control laws, are identified using system identification techniques. Moreover, the estimated models and the recorded dynamics are compared and contrasted. Finally, the Root Locus and Optimization-Cased Control System Design are implemented, making way for the tuning of the gains of the Piccolo II with regard to longitudinal movement. Significant improvement in the performance of the UAV Arcturus T-15 is also shown.
Description
Keywords
Piccolo II Ajuste dos ganhos Identificação de sistemas UAV Leis de controlo Dinâmica longitudinal Tuning of gains System identification Control laws Longitudinal dynamics
Citation
Publisher
Academia da Força Aérea