Control law automatic design method based on long short-term memory network

According to the control law automatic design method based on the long-short-term memory network, the control law structure design problem is converted into the directed acyclic graph topological relation automatic search problem by referring to the experience of a network architecture search method in deep learning and utilizing the advantages of a recurrent neural network in the aspect of time sequence relevance mining; according to the method, automatic generation of a flight control law control structure is realized, automatic setting of parameters under a given control law structure is realized based on a genetic algorithm, the limitation that automatic setting of controller parameters can only be carried out by utilizing a heuristic algorithm for a known controller structure in current control law automatic optimization is overcome, the workload of manual design is reduced, and the design efficiency is improved. And the control effect under the complex design input condition is improved. 本发明一种基于长短期记忆网络的