TY - JOUR
T1 - Aerodynamic/reaction-jet compound control of hypersonic reentry vehicle using sliding mode control and neural learning
AU - Shou, Yingxin
AU - Xu, Bin
AU - Liang, Xiaohui
AU - Yang, Daipeng
N1 - Publisher Copyright:
© 2021 Elsevier Masson SAS
PY - 2021/4
Y1 - 2021/4
N2 - Considering the saturation of the aerodynamic surface (ADS) and the limitation of the actuator input, the paper investigates the aerodynamic/reaction-jet compound attitude control for the hypersonic reentry vehicle subject to poor aerodynamic maneuverability. The attitude control torque is preferentially allocated to the ADS. If the ADS is saturated, the remaining control torque is distributed to the reaction control system (RCS) through the control distribution algorithm to assist the ADS. For the dynamics uncertainty in the calculation of control torque, the terminal sliding mode (TSM) controller based on the back-stepping frame is designed to improve the robust performance and achieve the finite-time convergence. Furthermore, the predefined-time TSM controller is constructed with the online-data neural learning and the disturbance observer, which guarantee the predefined-time convergence and complete the effective approximation of system dynamics uncertainty. The stability of the attitude system is mathematically proved via Lyapunov function, while the simulation results are provided to show the effectiveness of the proposed controllers.
AB - Considering the saturation of the aerodynamic surface (ADS) and the limitation of the actuator input, the paper investigates the aerodynamic/reaction-jet compound attitude control for the hypersonic reentry vehicle subject to poor aerodynamic maneuverability. The attitude control torque is preferentially allocated to the ADS. If the ADS is saturated, the remaining control torque is distributed to the reaction control system (RCS) through the control distribution algorithm to assist the ADS. For the dynamics uncertainty in the calculation of control torque, the terminal sliding mode (TSM) controller based on the back-stepping frame is designed to improve the robust performance and achieve the finite-time convergence. Furthermore, the predefined-time TSM controller is constructed with the online-data neural learning and the disturbance observer, which guarantee the predefined-time convergence and complete the effective approximation of system dynamics uncertainty. The stability of the attitude system is mathematically proved via Lyapunov function, while the simulation results are provided to show the effectiveness of the proposed controllers.
KW - Control allocation
KW - Hypersonic reentry vehicle
KW - Neural network
KW - Predefined-time terminal sliding mode
KW - Reaction control system
KW - Terminal sliding mode
UR - https://www.scopus.com/pages/publications/85102557277
U2 - 10.1016/j.ast.2021.106564
DO - 10.1016/j.ast.2021.106564
M3 - 文章
AN - SCOPUS:85102557277
SN - 1270-9638
VL - 111
JO - Aerospace Science and Technology
JF - Aerospace Science and Technology
M1 - 106564
ER -