论文部分内容阅读
基于有限集统计理论的概率假设密度滤波算法运用于多目标跟踪时,不再考虑数据关联问题,突破了传统的跟踪方法。但该滤波公式在非线性条件下没有解析解,在非线性高斯条件下提出了基于无迹变换的概率假设密度滤波算法,实现了算法在强杂波环境下的多目标跟踪。仿真实验比较了该算法与基于粒子滤波的概率假设密度滤波算法的跟踪性能,验证了该算法的跟踪性能和精度。同时分析指出了此算法的不足,以及结合无迹变换与粒子滤波的概率假设密度滤波算法的改进研究方向。
Probability Hypothesis Density Filtering Algorithm Based on Finite Set Theory of Statistics When used in multi-target tracking, the problem of data association is no longer considered and the traditional tracking method is broken. However, this filter has no analytical solution under non-linear conditions. Based on the non-linear Gaussian condition, a probability hypothesis density filtering algorithm based on no-trace transformation is proposed to realize the multi-target tracking under strong clutter. The simulation experiment compares the tracking performance of the algorithm with the probability hypothesis density filter algorithm based on particle filter, and verifies the tracking performance and accuracy of the algorithm. At the same time, the disadvantages of this algorithm are pointed out, and the improvement of probability density filtering algorithm combining with no-trace transform and particle filter is pointed out.