Abstract
A sub-optimal closed loop controller is proposed for the regulation of a dynamical system, such as a rocket, which uses a strictly bounded non-throttleable control. The system as well as measurements on its states are subject to random disturbances. With the assumption of a fixed constraint on a statistic of the terminal error, the controller attempts to minimize the total fuel consumption. The Control law is prestructured in the form u= i ω sgn k'x̂, where i ω is an indicator function which is unity during control action and zero elsewhere, and where k is a time-varying feedback gain vector. A solution to the problem consists of the computation of i ω and k as a function of a priori known statistics of the measurements, and the minimum cost is simply the integrated value of the value of the i ω indicator functions over the fixed time history of the vehicle's motion.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call