A modified version of the original differential dynamic programming (DDP) algorithm for unconstrained discrete optimal control problems is described. This version, which differs from the original by a second-order term which vanishes if the dynamics are linear, is proven to be a stagewise implementation of the classical Newton's method.
No takes yet. Share an insight, caveat, or question.
J. F. A. De O. Pantoja (1988) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: