| 基于Boltzmann机的NJPDA优化算法 |
| |
| 引用本文： | 徐永红,范跃华,辛大欣.基于Boltzmann机的NJPDA优化算法[J].西安工业学院学报,1999,19(4). |
| |
| 作者姓名： | 徐永红 范跃华 辛大欣 |
| |
| 作者单位： | 西安工业学院计算机科学与工程系!陕西西安710032 |
| |
| 基金项目： | 兵器工业预研基金!G9602 |
| |
| 摘 要： | 通过对联合概率数据关联的性能特征进行分析后认为，数据关联可归结为一类约束组合优化问题． 在此基础上，应用随机神经网络Ｂｏｌｔｚｍａｎｎ 机的组合优化求解策略，结合改进的模拟增益退火方法，引入了机动多目标快速随机神经联合概率数据关联组合优化算法（ＦＳＮＪＰＤＡ） ，改进后的算法克服了传统ＪＰＤＡ存在的计算组合爆炸现象，提高了ＮＪＰＤＡ的收敛速度．
|
| 关 键 词： | 神经网络 数据融合 数据关联 多目标跟踪 模拟退火 |
Optimun algorithm of NJPDA based on boltzmann machine |
| |
| XU Yong-hong,FAN Yue-hua,XIN Da-xin.Optimun algorithm of NJPDA based on boltzmann machine[J].Journal of Xi'an Institute of Technology,1999,19(4). |
| |
| Authors: | XU Yong-hong FAN Yue-hua XIN Da-xin |
| |
| Abstract: | In this paper,the properties of the joint probabilistic data association (JPDA) are analyzed,and the data association of multi maneuvering targets is reduced to be a sort of constraint combinatorial optimization problem. Based on Boltzmann machine and simulated gain annealing,an algorithm called fast stochastic neural joint probabilistic data association (FSJPDA) is presented. The simulations show that the computation combinatorial explore has been solved,and performance of the NJPDA has been improved. |
| |
| Keywords: | neural network data fusion data association multi target tracking simulated annealing |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
