The aim of this paper is to study the use of a genetic algorithm (CA) to optimise the trajectory of a conventional two-stage launcher The equations of motion of this lack analytical solutions, and the number of adjustable parameters is large enough the use of some non-traditional optimisation method becomes necessary. Two missions are considered first, to reach the highest possible stable, circular Low Earth (LEO); and second, to maximise the speed of a tangential escape trajectory. In this three variables are tuned and optimised by the CA in order to satisfy mission while maximising the target function. The technical characteristics and limitations launcher are taken into account in the mission model, and a fixed payload assumed. A variable mutation rate helps expand the search area whenever the Of Solutions becomes uniform, and is shown to accelerate convergence of the CA in cases. The obtained results are in agreement with technical specifications and obtained in the past (C) 2009 Elsevier Inc. All rights.