Abstract
Iterative dynamic programming (IDP) introduced by Luus has been recognized as a powerful method for solving a wide variety of nonlinear optimal control problems. Based on backward stagewise operation, the original IDP computational scheme performs the evaluation of the optimal control associated with each state node in a backward manner, starting from the last stage. In this paper, we present an improved scheme for carrying out the IDP computation which yields the same result as the backward IDP. The proposed scheme performs the control evaluation in a forward‐backward manner and thus can save computing time by eliminating the determination of optimal controls for inaccessible state nodes.
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