The control strategy of Lunar landing craft avoiding obstacles with minimum fuel consumption

Abstract

A comprehensive evaluation method and its terminal control strategy based on a simulated annealing algorithm are presented in this paper, which have application in avoiding the obstacles for lunar soft landing. Unlike many fuel-irrelevant researches or some classical strategies only concluding the routes, an innovative strategy considering the minimum fuel consumption and the obstacle avoidance strategy together is proposed, in view of a dynamics theory for minimum fuel consumption and computer simulation. The combined process consists of a coarse obstacle avoidance and a precise one, and in this paper, we take Chang-e’3 for example—all the following data and craft structures are based on Chang-e’3. Firstly, in order to optimize the algorithm for avoiding obstacles during the landing process, we need to create an index system to evaluate the degree of safety of the whole area divided into several sub-blocks. In the coarse obstacle avoidance section, at first, the digital picture taken from an elevation of 2400 m above the surface is divided into 100 m×100 m sub-blocks, and “the highest elevation”, “the lowest elevation”, “the minimum slope angle” are set as three indicators evaluating the landing safety of each area. The safety greyscale image can be drawn and the safety degree can be evaluated quantitatively by Cauchy membership function. Secondly, we develop a control scheme with high computational efficiency and apply it to select the safest landing area automatically. Taking the computational simulated trajectory into consideration, we adapt the simulated annealing algorithm to obtain the better landing area, in the process of which, a nonlinear programming model is introduced with the object function of minimum fuel consumption. With the parameters of the simulated annealing algorithm—the cooling coefficient a =0.999 and the terminal temperature E =10 - 30, the program nicely finishes with a high time efficiency. The entire operation of trajectory simulating, the safest landing area analyzing, the obstacle avoidance, the descent stages transforming and the comprehensive consideration of minimum fuel consumption are all based on computational simulation. Thirdly, according to the algorithm simulation, the landing trajectory including a specific landing coordinate is obtained. It takes 25.7 s for Chang’e-3 to complete the soft landing of coarse obstacle avoidance, with 63.86 kg fuel consumption. We also simulate with a given change of the initial conditions, in order to test and verify the validity of the algorithm, and results turn out to be very desirable. Similarly, the optimal control strategy for precise obstacle avoidance is also obtained, and in this section, the time cost and fuel consumption are both negligibly small. Finally, we focus on the error rate of the algorithm and provide a method to diminish the error rate—by adjusting weights of the three indicators which are used to evaluate the safety of sub-blocks. The algorithm can better deal with the effect of topographic diversity.