目标数未知时基于粒子滤波的多目标TBD方法

摘要

Since it is assumed that the number or the maximum number of possible targets is known in the current particle filter based track-before-detection (TBD) algorithms,and adjacent weak targets can not be detected effectively.Thus,this paper proposes a novel multi-target TBD algorithm based on particle filter and successive-target-cancellation (PF-STC).This novel algorithms deals with the multiple target tracking and detection with no prior information of the number of targets by simplifying K-dimension joint detection as multiple independent target detection.Compared with previous particle filter based multi-target TBD algorithms,the new proposed algorithm has no need for the prior information of the target number,overcomes the difficulty of the weaker target detection when it is close to the stronger target,and reduces algorithm complexity.The simulation results show that the new proposed algorithm can effectively improve the detection performance of multiple weak targets.%针对现有粒子滤波微弱多目标检测前跟踪(TBD)算法要求目标数目或者目标最大数目已知,且无法对邻近微弱目标有效检测的不足,提出了一种基于粒子滤波和目标相继消除(PF-STC)的多目标TBD算法.该算法通过将多目标状态的联合搜索过程简化为多个独立的单目标检测过程,实现数目未知的多目标跟踪和检测.与现有粒子滤波多目标TBD算法相比,新算法克服了现有方法在较弱目标接近较强目标时出现的检测困难,并降低了算法复杂度,能对数目未知的微弱多目标进行有效检测.