A Redundant Fault-tolerant Aviation Control System Based on Deep Neural Network

The paper is mainly focused on the redundant fault-tolerant control of control system. It is to solve the problem in existing fault-tolerant models with low reliability and high fault-tolerant failure rate by using deep neural network. The existing fault-tolerant models are impossible to predict the faults that will occur. In this paper, the overall framework structure of the redundant fault-tolerant system is designed. In terms of fault-tolerant control, the idea of hardware redundancy is adopted to build the double redundant controllers and the deep neural network is used for prediction. Then the implementation of the specific scheme is carried out for the actual scene of the agent tracking control.