Advantages of Differential Dynamic Programming Over Newton''s Method for Discrete-time Optimal Control Problems

摘要: Differential Dynamic Programming (DDP) and stagewise Newton''s method are both quadratically convergent algorithms for solving discrete time optimal control problems. Although these two share many theoretical similarities, they demonstrate significantly different numerical performance. In this paper, we will compare analyze in detail derive another algorithm which is a combination of the DDP method. This new second-order plays key role explanation differences between The detailed algorithmic structural three their impact on performance be discussed explored. Two test problems with various dimensions solved by presented. One nonlinear problem demonstrates that can as much 28 times faster than comparsion indicates numerically superior to