滑翔飞行器速度约束条件下制导律设计OACSTPCD

A neural network-based guidance method is proposed for the guidance problem of glide vehicles under velocity constraints.First,a trajectory optimization model for glide vehicles is established,and the trajectory optimization method based on the Gauss pseudospectral method and the guidance law model based on the neural network are proposed.Afterwards,in order to further meet the velocity constraints,a strategy for velocity control is proposed.Finally,the optimal control quantity obtained by the Gauss pseudospectral method is discretized as the tag data,which are used to train the back propagation(BP)neural network.The trained network is then embedded into the simulation model for numerical calculation,and the obtained results are compared with those obtained by the fixed control profile and Gauss pseudospectral method.The simulation results show that the guidance method proposed in this paper can meet the velocity constraints.It has the ability of online implementation and good engineering applications,which have certain reference significance for the guidance law design for such glide vehicles.