摘要
为提高基于滤波的多目标跟踪方法的性能,提出了一种多伯努利平滑方法。该方法由前向滤波和反向平滑两部分组成,前向滤波采用势平衡多目标多伯努利滤波,反向平滑利用多伯努利概率密度近似多目标平滑状态的概率密度,实现多目标平滑状态概率密度的反向递推计算。仿真结果表明,与滤波相比,多伯努利平滑对目标数量和目标状态的估计精度都有显著提高。
In order to improve the capability of multi-target tracking filter, a multi-Bernoulli smoother is proposed. This smoother consists of forward filtering followed by backward smoothing. The forward filtering is accomplished by the cardinal- ity-balanced multi-target multi-Bernoulli filter. For the backward smoothing, the backward recursion of the smoothed multi- target probability density is achieved by using muhi-Bernoulli approximation. Simulation results show that the proposed smoother improves the estimation accuracy of target number and target states over the filter.
出处
《数字通信》
2014年第2期8-11,共4页
Digital Communications and Networks
关键词
多目标跟踪
多伯努利
滤波
平滑
multi-target tracking, muhi-Bernoulli, filter, smoother
