航空学报 > 2001, Vol. 22 Issue (1): 54-56

引入神经网络的交互式多模型算法

陈利斌, 佟明安   

  1. 西北工业大学电子工程系 陕西西安 710072
  • 收稿日期:1999-10-20 修回日期:2000-01-30 出版日期:2001-02-25 发布日期:2001-02-25

INTERACTING MULTIPLE MODEL ALGORITHM WITH NEURAL NETWORKS

CHEN Li-bin, TONG Ming-an   

  1. Department of Electronics Engineering, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:1999-10-20 Revised:2000-01-30 Online:2001-02-25 Published:2001-02-25

摘要: 在交互式多模型算法中引入神经网络算法以改进目标跟踪的精度。利用神经网络算法对基于机动目标“当前”统计模型的均值和方差自适应滤波算法进行修改,提高该算法的性能,然后采用交互作用多模型算法跟踪机动目标,提高了机动目标的跟踪精度。

关键词: 交互式多模型, 神经网络, 机动目标, “当前”统计模型

Abstract: The interacting multiple model(IMM) algorithm has been shown to be one of the most effective estimation algorithms in the field of maneuvering target tracking. “Current”statistical model filtering algorithm is modified by neural networks and then adopted in IMM algorithm.The tracking performance for maneuvering target can be improved.The results of the modified IMM algorithm are evaluated through Monte Carlo simulation.

Key words: IMM, neur al net wo rks, m aneuv ering tar get, “curr ent”sta tistical model