| 引用本文: | 李振兴,刘进忙,李松,李延磊.一种改进的群目标自适应跟踪算法[J].哈尔滨工业大学学报,2014,46(10):117.DOI:10.11918/j.issn.0367-6234.2014.10.020 |
| LI Zhenxing,LIU Jinmang,LI Song,LI Yanlei.An improved adaptive tracking algorithm for group targets[J].Journal of Harbin Institute of Technology,2014,46(10):117.DOI:10.11918/j.issn.0367-6234.2014.10.020 |
|
| 摘要: |
| 为提高对群目标在机动情况下的跟踪性能,提出一种改进的群目标自适应跟踪算法．在群质心状态估计中,在修正“当前”统计模型的基础上,利用群质心的速度预测和速度估计的偏差进行过程噪声方差自适应调整,并引入强跟踪滤波中的渐消因子,实时调节群质心的状态预测协方差．在扩展状态估计中,将其对应的椭圆面积预测值和估计值的偏差以及偏差变化率作为模糊输入量,采用模糊推理法自适应输出扩展状态的预测参数．此外,提供了群目标分裂机动的判决方法．仿真结果表明,与现有方法相比,本文算法增强了对群目标在突发机动时的自适应跟踪能力,并能有效检测出群的分裂机动． |
| 关键词: 群目标 质心状态 扩展状态 强跟踪滤波 模糊推理 分裂机动 |
| DOI：10.11918/j.issn.0367-6234.2014.10.020 |
| 分类号:TN953 |
| 基金项目:国家自然科学青年基金项目(61102109);陕西省自然科学基金项目(2010JM8013);空军工程大学防空反导学院“研究生科技创新基金”项目(HX1112). |
|
| An improved adaptive tracking algorithm for group targets |
|
LI Zhenxing, LIU Jinmang, LI Song, LI Yanlei
|
|
(Air and Missile Defense College, Air Force Engineering University, 710051 Xi’an, China)
|
| Abstract: |
| In order to improve tracking performance of the approach, a new adaptive tracking algorithm of group maneuvering targets was presented. In the estimation of group centroid kinematic state, the deviation between the prediction value and estimation value of centroid speed was used to adjust the covariance matrix of process noise based on modified current statistical model, and a fading factor of strong tracking filter was used to adjust the state-estimation error covariance adaptively. In the estimation of group extension state, the prediction parameter of extension was calculated by using a fuzzy reasoning method, which had taken the deviation between the prediction value and estimation value of the corresponding elliptical area and the change ratio of deviation as the input of the fuzzy controller. Lastly, a method to judge split-off maneuvering of group targets was offered. Simulation results show that, compared with the existing methods, the proposed algorithm can obtain a better adaptive tracking performance in maneuvering scenarios, and detect the split-off maneuvering effectively. |
| Key words: group targets centroid kinematic state extension state strong tracking filter fuzzy reasoning split-off maneuvering |
