| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 3.24 MB | Adobe PDF |
Advisor(s)
Abstract(s)
A presente dissertação visa a análise do problema de orientação de mísseis intercetores
pela abordagem dos princípios de controlo ótimo de sistemas dinâmicos. As medições de
posição são incertas, devido à falta de dados ou a erros a estes associados, especialmente
quando se trata de alvos que se movem a velocidades elevadas. É necessário também
garantir que o controlo do intercetor é efetuado de forma célere, eficaz e dentro dos seus
constrangimentos de manobrabilidade. Para a estimação de posições foi estudado um
algoritmo de filtragem de Kalman adaptativa com base na inovação, aplicando-o a
trajetórias que podem ser altamente ruidosas. Quanto ao controlo, escolheram-se
princípios de controlo integral multivariável com base na abordagem de controlo LQR,
que foram aplicados ao modelo de um intercetor face a vários tipos de trajetória. Em
ambos os casos os resultados obtidos mostraram-se interessantes no que diz respeito aos
requisitos de interceção. O filtro de Kalman adaptativo mostrou-se capaz de reduzir
significativamente o efeito dos erros do sistema, apresentando uma trajetória mais
próxima à real. A flexibilidade de implementação graças à adaptabilidade do filtro é outro
ponto a seu favor. O controlador LQR adaptado conseguiu intercetar o alvo com sucesso
para os vários casos estudados, notando-se uma ligeira alteração no desempenho na
aproximação ao alvo quando este executa manobras. O controlador apresentou uma
robustez adequada, mantendo as manobras do intercetor dentro de parâmetros
predefinidos. Os resultados das simulações mostraram que o filtro de Kalman adaptativo
é uma ferramenta viável na estimação de posições em sistemas mesmo em casos de
incertezas de medição e do modelo, bem como de perturbações no ambiente. De forma
similar o controlador LQR adaptado mostra-se capaz de guiar um intercetor de forma
rápida e eficaz, apresentando também boas qualidades de robustez. A utilidade do
método foi também verificada e apresenta-se como uma possível combinação a utilizar
para a interceção de potenciais ameaças.
This dissertation aims to study the problem of missile interceptor guidance following the principles of optimal control of dynamic systems. Position measurements are uncertain due to missing data or associated errors, especially when dealing with high-speed targets. It is also necessary to guarantee that the interceptor control is achieved quickly and effectively, within the manoeuvrability constraints. An adaptive innovation based Kalman filtering algorithm was studied to tackle the estimation problem, applying the algorithm to trajectories that may be highly noisy. For the control problem, Multivariate integral control principles based on the LQR control approach were chosen and applied to the model of an interceptor faced with various target flight paths. On both cases the obtained results were found to be interesting with respect to interception requirements. The adaptive Kalman filter proved capable of significantly reducing the effects of system errors, displaying a trajectory that was closer to the real one. The filter also showed great implementation flexibility due to its adaptability. The extended LQR controller was able to successfully intercept the target for all the studied cases, with a slight change in performance when the target is manoeuvring. The controller showed an adequate robustness while maintaining the interceptor manoeuvres within the predefined parameters. The simulation results showed that the adaptive Kalman filter was a viable tool in systems position estimation even in case of model and measurement uncertainties and environmental disturbances. In a similar fashion, the LQR controller proved itself capable of guiding an interceptor quickly and effectively while showing good robustness. The usefulness of the method was also verified and presented as a possible combination to be used for potential threat interception.
This dissertation aims to study the problem of missile interceptor guidance following the principles of optimal control of dynamic systems. Position measurements are uncertain due to missing data or associated errors, especially when dealing with high-speed targets. It is also necessary to guarantee that the interceptor control is achieved quickly and effectively, within the manoeuvrability constraints. An adaptive innovation based Kalman filtering algorithm was studied to tackle the estimation problem, applying the algorithm to trajectories that may be highly noisy. For the control problem, Multivariate integral control principles based on the LQR control approach were chosen and applied to the model of an interceptor faced with various target flight paths. On both cases the obtained results were found to be interesting with respect to interception requirements. The adaptive Kalman filter proved capable of significantly reducing the effects of system errors, displaying a trajectory that was closer to the real one. The filter also showed great implementation flexibility due to its adaptability. The extended LQR controller was able to successfully intercept the target for all the studied cases, with a slight change in performance when the target is manoeuvring. The controller showed an adequate robustness while maintaining the interceptor manoeuvres within the predefined parameters. The simulation results showed that the adaptive Kalman filter was a viable tool in systems position estimation even in case of model and measurement uncertainties and environmental disturbances. In a similar fashion, the LQR controller proved itself capable of guiding an interceptor quickly and effectively while showing good robustness. The usefulness of the method was also verified and presented as a possible combination to be used for potential threat interception.
Description
Keywords
Controlador Lqr Controlo Ótimo Defesa Territorial Filtragem de Kalman Adaptativa Interceção Por Míssil