A multiple-shooting differential dynamic programming algorithm. Part 2: Applications. (August 2020)

• Record Type: Journal Article
• Authors: Pellegrini, Etienne; Russell, Ryan P.
• Is Part Of: Acta astronautica. Volume 173(2020)
• Page Start: 460
• Page End: 472
• Publication Date: 2020-08
• Subjects: Optimal control -- Trajectory optimization -- Differential dynamic programming -- Multiple-shooting -- Augmented Lagrangian; Astronautics -- Periodicals; Outer space -- Exploration -- Periodicals; Astronautics; Periodicals; 629.405
• DOI: 10.1016/j.actaastro.2019.12.038
• Languages: English
• ISSNs: 0094-5765

Abstract

Abstract: A Multiple-Shooting Differential Dynamic Programming Algorithm is applied to a variety of constrained nonlinear optimal control problems, including classic benchmark problems, as well as a robotic arm problem and sensitive spacecraft trajectory optimization problems. The multiple-shooting extension presented in Part 1 is validated, as well as the Powell, Hestenes, and Rockafellar approach used in the treatment of equality and inequality path and terminal constraints. The results for example applications demonstrate the applicability of the algorithm to a variety of trajectory optimization problems. The advantages of the multiple-shooting approach over the single-shooting algorithm is evident in problems with high sensitivity. In particular, convergence of the multiple-shooting algorithm is demonstrated for complex spacecraft trajectory problems that are intractable using the single-shooting formulation. Highlights: The multiple-shooting differential dynamic programming algorithm is validated. The multiple-shooting approach is effective for general optimal control problems. The sensitivities of each subproblem are reduced with multiple-shooting. The new algorithm can be used to solve complex and sensitive problems robustly. The Augmented Lagrangian approach to inequality constraints is effective.