神经网络在广义经典分配问题中的应用

Application of Neural Network in Generalized Classical Assignment Problem

摘要:

在两节点分布式多传感器系统中,一些航迹关联算法可以化为广义经典分配问题。广义经典分配问题是一个组合优化问题。当目标数目较多时,很难得到问题的最优解,而且其计算量容易呈现指数爆炸现象。文章提出了用Hopfield神经网络和平均场网络解决此问题的方法。仿真结果表明,采用文章提出的人工神经网络模型求解广义经典分配问题,不仅使航迹关联具有较高的关联正确率,而且计算时间不会出现指数爆炸现象。仿真结果还表明,平均场网络相比Hopfield神经网络更易于得到问题的最优解。

Abstract:

In a two-node distributed multisensor system, some algorithms of track correlation can be transformed to a generalized classical assignment problem, and the generalized classical assignment is a combinational optimization problem. When the number of targets is large, it is very hard to obtain the optimum solution and its computing burden often increases exponentially. The model of Hopfield neural network and mean-field network were used in this paper to solve the problem. The simulation results illustrate that using the neural networks to solve the generalized classical assignment problem not only makes the track association correct percent high, but also cannot increase the computing time exponentially with the number of targets. The experiments also show that the optimum solution can be easily obtained using mean-field neural network than using Hopfield neural network.