Prediction method of spacecraft flight capability in atmospheric entry phase based on Gaussian process regression

Trajectory optimization technology is one of the key technologies in the study of atmospheric entry phase. How to evaluate the performance parameters of the entry trajectory under the conditions of complicated atmospheric entry dynamics, different spacecraft design parameters and multiple constraints of the entry process is an important problem in the study of trajectory design. Thus, the maximum flight range of the atmospheric entry phase represented by the two-dimensional landing point corridor is taken as the performance index. Aiming at the problem that the traditional trajectory optimization method has a large amount of calculation, a fast prediction method of the flight capacity of the atmospheric entry phase based on the Gaussian process regression (GPR) is proposed to mine the mapping relationship between the initial trajectory parameters of the spacecraft and the characteristic parameters of the trajectory envelope. The method avoids complex dynamic modeling and large-scale iterative optimization process when solving the maximum range of spacecraft. By using the proposed method, the maximum range of entry trajectory of more than 1 000 groups of different entry scenarios is predicted rapidly, and the predicted results are used to evaluate the flight ability of entry spacecraft, thus providing reference for solving the engineering problems related to atmospheric entry. © 2020, Editorial Office of Systems Engineering and Electronics. All right reserved.