Low-Thrust Fuel-Optimal Trajectory Optimization Using Different Initial Guess Methods

Xing Cheng,Xiaoning Shi,Naigang Cui
DOI: https://doi.org/10.1109/iceice.2012.638
2012-01-01
Abstract:The feasible and application scope of two initial guess methods used for low-thrust fuel-optimal trajectory optimization are investigated in this paper. Indirect method is chosen here to solve the optimal trajectory, and the resulting NLP is solved by SQP algorithm. Then two initial guess methods are introduced, one is ad joint control transformation, the other is first order Taylor series method. After deriving the two initial guess methods, numerical examples are given. The results demonstrate that both initial guess methods are suitable for low-thrust fuel-optimal trajectory optimization, furthermore, for short-duration missions, the Taylor series method is better than ACT, while for the long-duration missions, the ACT is slightly more effective, based on this point, some complex missions could be solved.
What problem does this paper attempt to address?