Open Access

Bernd Dachwald, Andreas Ohndorf

• Searching optimal continuous-thrust trajectories is usually a difficult and time-consuming task. The solution quality of traditional optimal-control methods depends strongly on an adequate initial guess because the solution is typically close to the initial guess, which may be far from the (unknown) global optimum. Evolutionary neurocontrol attacks continuous-thrust optimization problems from the perspective of artificial intelligence and machine learning, combining artificial neural networks and evolutionary algorithms. This chapter describes the method and shows some example results for single- and multi-phase continuous-thrust trajectory optimization problems to assess its performance. Evolutionary neurocontrol can explore the trajectory search space more exhaustively than a human expert can do with traditional optimal-control methods. Especially for difficult problems, it usually finds solutions that are closer to the global optimum. Another fundamental advantage is that continuous-thrust trajectories can be optimized without an initial guess and without expert supervision.