计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (22): 235-242.DOI: 10.3778/j.issn.1002-8331.1605-0371

• 工程与应用 • 上一篇    下一篇

强跟踪稀疏网格滤波在水下目标跟踪中的应用

王  越,徐正生,于  飞,陈斐楠   

  1. 哈尔滨工程大学 理学院,哈尔滨 150001
  • 出版日期:2017-11-15 发布日期:2017-11-29

Application of strong tracking sparse grid filter in underwater target tracking

WANG Yue, XU Zhengsheng, YU Fei, CHEN Feinan   

  1. School of Science, Harbin Engineering University, Harbin 150001, China
  • Online:2017-11-15 Published:2017-11-29

摘要: 针对水下目标发生强机动的情况,为解决稀疏网格求积滤波(Sparse Grid Quadrature Filter,SGQF)精度下降,甚至发散的问题,提出了强跟踪稀疏网格求积滤波(Strong Tracking Sparse Grid Quadrature Filter,STSGQF)算法。SGQF能够在保证滤波精度的情况下,大大降低运算成本。在此基础上引入强跟踪滤波(Strong Tracking Filter,STF),STSGQF不仅保留了SGQF跟踪精度高、运行时间短的优点,还提高了算法的鲁棒性,同时,也解决了STF需要求解Jacobian矩阵的问题。通过仿真实验验证了STSGQF的有效性。

关键词: 水下目标跟踪, 高斯-埃尔米特求积滤波, 稀疏网格, 强跟踪滤波

Abstract: A Strong Tracking Sparse Grid Quadrature Filter(STSGQF) is proposed to overcome the problem that Sparse Grid Quadrature Filter(SGQF) decreases in accuracy, even diverges when the underwater target moves with power maneuverability. The sparse grid is avaible to reduce the SGQF’s computation cost, significantly, while the accuracy lowers slightly. STSGQF where the Strong Tracking Filter(STF) is introduced into SGQF not only keeps the advantages of high accuracy and shorter running time, but also improves the robustness of the algorithm. Meanwhile, STSGQF need not to calculate Jacobian matrix. Simulation results show the effectiveness of STSGQF.

Key words: underwater target tracking, Gauss-Hermite quadrature filter, sparse grid, strong tracking filter