探测点序列的最优化与简化计算方法

Optimization and Simplified Calculation Method of Detection Point Sequence

摘要:

在搜索状态建模和求解一阶搜索状态方程的特征迹线解的基础上,建立了对随机运动进行离散时间探测的发现概率最优控制模型,结合动态规划原理给出一种最优探测点序列的逼近算法,并给出了短时计算的算法简化形式。在满足一阶搜索状态方程的随机恒速目标条件以及有限指数探测函数条件下,将给出的算法及其简化形式应用到算例。算例表明,当随机恒速运动目标初始位置和速度均服从圆正态分布时,该算法及其简化形式均能够由任意给定的初始探测点序列优化收敛到满足精度要求的最优探测点序列。

Abstract:

On the basis of the first order search state model,the optimal control model of the probability to find a randommotion target with discrete time detection was established. Then, an approximation algorithm and the simplified form ofshort time of the algorithm for the optimal detection point sequences were given based on the dynamic programming princi.ple. The algorithm and its simplified form of short time were applied to the example under the condition of satisfying of thefirst order search equation of the random constant speed target condition and the finite exponential detection function. Thesimulation results showed that when the initial location and velocity of the random constant velocity target obeyed the cir.cular normal distribution, the algorithm and its simplified form of short time could be converged by any given initial detec.tion point sequence to the optimal detection point sequence satisfying the accuracy requirement.