Predictor-corrector reentry guidance based on online model identification

Aiming at the improvement of autonomy, adaptability and robustness of the reentry guidance technology, a new predictor-corrector reentry guidance law based on online model identification is proposed under the background of the new generation of reusable aircraft. On the basis of equilibrium glide condition, a feasible control envelope is established to satisfy the reentry constraints. The corresponding relationship between the feasible control profile and the feasible range capability is analyzed and proved, thus the initial trajectory generator (ITG) based on the Gauss-Newton method is designed. The generator does not rely on ground equipment, and has certain autonomy for the circumstance of task adjustment and emergency situation in flight process. Aiming at nonlinearity and uncertainty of the reentry model, an adaptive trajectory corrector based on sliding mode guidance law and RBF neural network which is used to approximate the nonlinear function is proposed. The control profile is adaptively modified by the guidance to achieve real-time correction of longitudinal deviation. Besides, the closed-loop control law is enforced by real-time feedback of the aircraft state, and the guidance command is restricted so as not to depend on the equilibrium gliding condition. The lateral guidance adopts the bank angle reversing strategy to decrease the cross-range progressively, so that the number of reversal is controllable. The simulation results show that the guidance method has high accuracy and good robustness in the presence of multiple disturbances and deviations.