The artillery firing precision plays an important role in the war and it's hard to describe the projectile trajectory in a mathematical model. In this paper, the neural network is used to build the artillery ballistic model for range prediction and Particle Swarm Optimization (PSO) is applied to optimize the initial weight and bias to accelerate the training speed. Besides, some firing data from one middle-caliber artillery are utilized in orthogonal array to reduce the experimental runs. The result shows that the proposed method has the faster training speed and better precision of range prediction than the traditional neural networks and proves to build quickly a suitable artillery ballistic model in less firing data without the complicated mathematical equations.
Chen Yi-Wei, National Defense University, Taiwan
This paper is part of the ACSET2013 Conference Proceedings (View)
View / Download the full paper in a new tab/window