Optimal nonlinear control and estimation for a reusable launch vehicle during reentry phase
2008 16th Mediterranean Conference on Control and Automation, 2008•ieeexplore.ieee.org
In this paper a nonlinear optimal controller has been designed for aerodynamic control
during the reentry phase of the reusable launch vehicle (RLV). The controller has been
designed based on a recently developed technique optimal dynamic inversion (ODI). For full
state feedback the controller has required full information about the system states. In this
work an extended Kalman filter (EKF) is developed to estimate the states. The vehicle (RLV)
has been has been consider as a nonlinear six-degree-of-freedom (6-DOF) model. The …
during the reentry phase of the reusable launch vehicle (RLV). The controller has been
designed based on a recently developed technique optimal dynamic inversion (ODI). For full
state feedback the controller has required full information about the system states. In this
work an extended Kalman filter (EKF) is developed to estimate the states. The vehicle (RLV)
has been has been consider as a nonlinear six-degree-of-freedom (6-DOF) model. The …
In this paper a nonlinear optimal controller has been designed for aerodynamic control during the reentry phase of the reusable launch vehicle (RLV). The controller has been designed based on a recently developed technique optimal dynamic inversion (ODI). For full state feedback the controller has required full information about the system states. In this work an extended Kalman filter (EKF) is developed to estimate the states. The vehicle (RLV) has been has been consider as a nonlinear six-degree-of-freedom (6-DOF) model. The simulation results shows that EKF gives a very good estimation of the states and it is working well with ODI. The resultant trajectories are very similar to those obtained by perfect state feedback using ODI only.
ieeexplore.ieee.org
以上显示的是最相近的搜索结果。 查看全部搜索结果