Control Allocation Reconstruction of Launch Vehicle Based on Neural Network

Abstract

After the single engine thrust reduction fault of the launch vehicle, if the conventional fixed control allocation method is used, the actual control torque generated by the engine is difficult to meet the expected torque generated by the basic controller. Therefore, this paper proposes a neural network based control allocation method. The method learns the pseudo-inverse allocation method, so that the actual control torque can well track the desired control torque. Finally, the proposed neural network control allocation method is simulated and verified. The results show the feasibility and effectiveness of the proposed method.