Evolutionary Neurocontrol: A Novel Method for Low-Thrust Gravity-Assist Trajectory Optimization

摘要

The combination of low-thrust propulsion and gravity assists to enhance deep-space missions has proven to be arnremarkable task. In this paper, we present a novelmethod that is based on evolutionary neurocontrollers. Themainrnadvantage in the use of a neurocontroller is the generation of a control law with a limited number of decisionrnvariables. On the other hand, the evolutionary algorithm allows one to look for globally optimal solutions morernefficiently than with a systematic search. In addition, a steepest-ascent algorithm is introduced that acts as arnnavigator during the planetary encounter, providing the neurocontroller with the optimal insertion parameters.rnResults are presented for a Mercury rendezvous with a Venus gravity assist and for a Pluto flyby with a Jupiterrngravity assist.