Neuro-optimal guidance control for lunar soft landing

Returning to moon has become a top topic recently. Many studies have shown that soft landing is a challenging problem in lunar exploration. The lunar soft landing in this paper begins from a 100km circular lunar parking orbit. Once the landing area has been selected and it is time to deorbit for landing, a ΔV burn of 19.4m/ s is performed to establish a 100 × 15km elliptical orbit. At perilune, the landing jets are ignited, and a propulsive landing is performed. A guidance and control scheme for lunar soft landing is proposed in the paper, which combines optimal theory with nonlinear neuro-control. Basically, an optimal nonlinear control law based on artificial neural network is presented, on the basis of the optimum trajectory from perilune to lunar surface in terms of Pontryagin's maximum principle according to the terminal boundary conditions and performance index. Therefore some optimal control laws can be carried out in the soft landing system due to the nonlinear mapping function of the neural network. The feasibility and validity of the control laws are verified in a simulation experiment.