There are two classes of problems in the general theory of optimal control: the problem of selecting a programmed control, and the problem of synthesis. Problems of the first class are solved during investigations of long-range planning, in work on preliminary calculations of the trajectories of rocket motions. Problems of optimal control synthesis always arise in real time control of an object. While there are fairly effective numerical methods for solving problems of the first class, as yet no efficient methods for solving synthesis problems have been found, although several attempts have been made to create them. The method proposed in this article is based on the calculation of the values of the derivative of Bellman's function at certain grid points in the phase space; this function is the vector of initial conditions for a certain system of differential equations whose solution determines the optimal transition from the chosen point to the end of the tube.
Numerical method for approximate synthesis of a local optimal control (rocket trajectory)
Eine numerische Methode fuer die approximative Synthese einer oertlichen optimalen Regelung
Cybernetics, New York ; 9 , 5 ; 869-874
1973
, 4 Quellen
Aufsatz (Zeitschrift)
Englisch
Approximate solutions to optimum climbing trajectory for rocket-powered aircraft
Engineering Index Backfile | 1959
|A study on the optimal rocket trajectory
AIAA | 1971
|