A Neural Network Adaptive Fault-Tolerant Control Method for Launch Vehicles with the Limited Faults

Haiyang Zhu,Yu Chen,Yansheng Wu,Yi Rong,Xudong Qin

doi:10.1051/jnwpu/20203830668

Abstract

To tolerate the limited faults such as thrust decline or actuator jamming of launch vehicles, an adaptive fault-tolerant control method based on radial basis function neural network (RBFNN) is proposed in this paper. The method is based on a limited faults dynamics model, and the baseline controller is designed based on the pole placement, using RBFNN to online identify and compensate the fault parameters and uncertain disturbances in the model. Then an adaptive fault-tolerant control law is designed based on Lyapunov theory. The simulation results show that the proposed adaptive control method can effectively ensure the attitude stability as well as control accuracy under the limited faults of launch vehicles, compared with the traditional PD control method.

Highlights

示 X 可行空间;且权值向量可行域 Ωf 满足条件 Ωf = { W^ | ‖W^ ‖ ≤ M,M ∈ R + } ,则 RBFNN 理想权值定义为 W∗T = arg min[ sup | f( x,d,t) - ^f( x,d,t) | ] , W^ ∈Ωf X∈Uc
To tolerate the limited faults such as thrust decline or actuator jamming of launch vehicles, an adaptive fault⁃tolerant control method based on radial basis function neural network ( RBFNN) is proposed in this paper
The method is based on a limited faults dynamics model, and the baseline controller is designed based on the pole placement, using RBFNN to online identify and compensate the fault parameters and uncertain disturbances in the model

Summary

Introduction

示 X 可行空间;且权值向量可行域 Ωf 满足条件 Ωf = { W^ | ‖W^ ‖ ≤ M,M ∈ R + } ,则 RBFNN 理想权值定义为 W∗T = arg min[ sup | f( x,d,t) - ^f( x,d,t) | ] , W^ ∈Ωf X∈Uc Ìïu = u1 + u2 íïu1 = M0( qd - kvė - kp e) + C0q + K0q (18) îïïu2 = - M0^f( x,W^ ) 式中, ^f( x,W^ ) a 为 RBFNN 对实际 f( x) 辨识值。表 ΦT( x) W􀮃ETPx + ηTETPx) ] + 1 tr( W􀮃TW􀮃) = γ - 1 xTQx + φT( x) W􀮃ETPx + ηTETPx + 1 tr( W􀮃TW􀮃)

Results

Conclusion