基于无味粒子滤波和交互多模型算法的多机动目标跟踪.pdfVIP

22 6 ( ) Vo122 No6 2008 12 Journalof JiangsuUniversity of Science and Technology(NaturalScience Edition) ec. 2008 基于无味粒子滤波和交互多模型算法的 多机动目标跟踪 何祖军, 尚明玲 ( , 212003) : . , IMM ( InteractingM ul tipleM odels) IMM , UPF (Unscented Particle Filter) IMM , UPF_IMM , IMM_JP A IMM , , . : ; ; IMM ; : TN953 : A : 1673- 4807( 2008) 06- 0048- 05 M ultitarget tracking based on unscented particle filter and interacting multiple m odels algorithm HE Zujun, SHANGM ingling ( School of Electronics and Information, Jiangsu University of Science and Technology, Zhenjiang Jiangsu 212003, China) Abstract: Glintnoise is a kind of nonGaussian noise. In order to mi prove tracking precision of multitarget under glint noise, UPF_IMM (Unscented Particle Filter Interacting Multiple Models) algorithm is proposed based on IMM algorithm in this paper. Also, nonlinear and nonGaussian filter Particle Filter(PF) is inte grated w ith IMM algorithm. Unscented PF(UPF) is applied to take the place of Kalman filter in IMM algo rithm. Furthermore, UPF_IMM is used to take theplaceof IMM in traditionaldata association algorithm IMM _JP A; and then multitarget tracking under glintnoise is resolved. Smi ulation results indicate the proposed algorithm can obviously enhance tracking precision. K ey w ord s: multitarget tracking; particle filter; IMM algorithm; glint noise 0 , , ,,. IMM ( InteractingMulti pleModels), (), ,., PF( Particle Filter) : 2008- 10- 09 : ( BG2006022) : ( 1964- ),, , , . Email: zjjskdhz@j yahoo. com. cn 6 ,: 49 [