论文部分内容阅读
在高斯混合多扩展目标PHD滤波的基础上,结合最新兴起的箱粒子滤波,提出一种基于区间分析的多扩展目标PHD滤波算法.采用大小可控的非零矩形区域来代替传统的多个点量测,这样可降低权值计算中对量测分布的要求.仿真对比实验表明,采用区间分析方法在保证近似于传统滤波精度的同时可降低计算复杂度,在目标数目估计及抗杂波干扰方面也具有较为突出的优势,并且可解决在目标靠近时由于不能正确给出子划分而造成的漏检问题.
Based on the Gaussian hybrid multi-extension target PHD filter and combined with the newly emerging box particle filter, a multi-extension target PHD filtering algorithm based on interval analysis is proposed, which uses non-zero rectangular area with controllable size to replace the traditional multiple point So as to reduce the requirement of measurement distribution in weight calculation.The simulation experiment shows that the interval analysis method can reduce the computational complexity while ensuring the accuracy of traditional filtering and the number of targets and the ability to resist clutter Also has a more prominent advantages, and can solve the problem in the absence of the target due to sub-division can not be correctly caused by the problem.