Abstract
This paper proposes that the deep neural network-based guidance (DNNG) law replace the proportional navigation guidance (PNG) law. This approach is performed by adopting a supervised learning (SL) method using a large amount of simulation data from the missile system with PNG. Then, the proposed DNNG is compared with the PNG, and its performance is evaluated via the hitting rate and the energy function. In addition, the DNN-based only line-of-sight (LOS) rate input guidance (DNNLG) law, in which only the LOS rate is an input variable, is introduced and compared with the PN and DNNG laws. Then, the DNNG and DNNLG laws examine behavior in an initial position other than the training data.
Highlights
Proportional navigation guidance (PNG) is one of the most widely known missile guidance laws in existence [1]
The action can be derived through the DNN-based guidance (DNNG) model for any state, and the mean absolute error (MAE) can evaluate the difference between the accelerations derived from the DNNG and from the proportional navigation guidance (PNG)
DNNG is compared with PNG, and with the reinforcement learning-based guidance (RLG) law presented in the previous paper [16], and this demonstrates its performance through the energy term
Summary
Proportional navigation guidance (PNG) is one of the most widely known missile guidance laws in existence [1]. This guidance law generates an acceleration command proportional to the line-of-sight (LOS) rate from the perspective of a missile looking at a target. This is easy to implement based on simple principles. Sci. 2020, 10, 7865 using feed-forward multi-layer perceptron (MLP) They demonstrated the efficiency of the proposed method by performing simulations on various initial input values. They demonstrated the efficiency thepresented proposed an ANN-based approach using MLP to replace showed that MLP-based guidance method by performing simulations on various initialPNG. It showed that MLP-based guidance can effectively In this propose a deep neural network-based replace the paper, PNG inwe a wide range of engagement scenarios. The initial values of the missile have random values within the presented range, and the simulation is terminated if the final distance of 2 m is satisfied
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
