A three-dimensional proportional guidance law based on RBF neural network

A new three-dimensional (3D) pure proportional guidance law based on radial basis function (RBF) neural network (PNGLRBF) design method was proposed. The proposed method was applied to decreasing the miss distance, which is mostly arisen from the fixed navigation rates of traditional proportional navigation guidance laws (TPNGLs). Firstly, using the engagement geometry formula between missile and aircraft, a typical 3D augment proportional guidance law was presented. Secondly, a simple performance index about miss distance was created, and then RBF neural networks were employed to tune the navigation rates on line to produce optimal over-load commands. Finally, two engagement scenarios were examined, and a comparison between TPNGLs and the proposed PNGLRBF was made, the simulation results show that the proposed the proposed 3D PNGLRBF can achieve ideal miss distance than TPNGLs.