关键词: 区域射击, 离散概率空间, 梯度下降法, 变分

Abstract: A gradient descent-based aiming point solving method for area firing in discrete probability space is presented to enhance the applicability of area firing method and solve the area firing aim points in the case of irregular distribution of targets. The target distribution probability in discrete probability space is described to adapt to irregular target distribution. The optimal intermediate function in discrete probability space is deduced using the variational method. The analytical gradient function of index is deduced by setting the norm of difference value between practical intermediate function and optimal intermediate function as index function. An aim point solving model is constructed based on gradient descent method. The geometric method, function-approximation method and gradient descent method are used to obtain the distributions of aiming points under the equivalent initial condition for different target distributions, respectively. The results show that the gradient descent method has better applicability, faster solving speed and higher degree of optimization for solving the aim-points configuration for area firing under the condition of irregular target distribution. Key

Key words: areafiring, discreteprobabilityspace, gradientdescentmethod, variation