Halo-to-halo cost optimal transfer based on CMA-ES

Author

Lian Yijun ; Jiang Xiaoyong ; Tang Guojian

Author_Institution

Coll. of Aerosp. Sci. & Eng., Nat. Univ. of Defense Technol., Changsha, China

fYear

2013

Firstpage

2468

Lastpage

2473

Abstract

A method is developed to implement the cost-optimal transfer between two halo orbits of the same family. The problem is recast as a three-body Lambert´s problem (TBLP), which is solved efficiently and reliably by a revised version of the algorithm initially proposed by Sukhanov and Prado. Six scalar optimization parameters are selected by analyzing the design variables of the problem and the characteristics of the iterated algorithm for solving the TBLP. The searching range of each optimization parameter is provided for speeding up the convergence of the evolution process. The performance index is the sum of velocity discontinuities at both ends of the two-impulse transfer orbit, and the cost-optimal result is obtained by using the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), which saves approximately 16% fuel as compared with the relevant work.

Keywords

control system synthesis; covariance matrices; evolutionary computation; iterative methods; optimisation; search problems; space vehicles; CMA-ES; TBLP; covariance matrix adaptation evolution strategy; design variables; evolution process; halo orbits; halo-to-halo cost optimal transfer; iterated algorithm; performance index; scalar optimization parameters; searching range; three-body Lambert problem; two-impulse transfer orbit; velocity discontinuities; Covariance matrices; Moon; Optimization; Orbits; Performance analysis; Space vehicles; Trajectory; CMA-ES; Covariance Matrix Adaptation Evolution Strategies; Halo-to-halo transfer; Three-body Lambert´s problem; optimization;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (CCC), 2013 32nd Chinese

Conference_Location

Xi´an

Type

conf

Filename

6639841