Optimal soft lunar landing based on differential evolution

Soft lunar landing optimization is a complex optimal control problem. Applying polynomial interpolation, control curves are expressed by N parameters which represent ordinates at N-Order Chebyshev polynomial´s roots. The states of the problem are determined by numerical integration. Then, the problem is translated to a nonlinear programming problem(NLP) whose decision vector is the parameters of the interpolation polynomials. It is solved by an efficient stochastic algorithm-differential evolution(DE) incorporated with constraints processing method. The algorithm is convenient to implement due to only a few parameters need to be set. In order to evaluate the algorithm, a scenario simulation is given. The results are compared with a direct transcription method in literature, and it shows that the solution of our algorithm is comparable to the counterpart.