Differential Dynamic Programming: An Optimization Technique for Nonlinear Systems

Abstract

<p> Differential Dynamic. Programming is a new method, based on Bellman's principle of optimality, for determining optimal control strategies for nonlinear systems. It has originally been developed by D.H.Jacobson. </p> <p> In this thesis a result is presented for a problem with saturation characteristics in nonlinearity solved by the Jacobson's approach. In the differential dynamic programming the principle of optimality is applied to the differential change in non-optimal cost due to small changes in state; variables instead of the cost itself. This results in modest memory requirements for its defining parameters and rapid convergence. </p>