一种加权邻域数据关联算法研究

Study on weighted neighbor data association algorithm

借鉴概率数据关联的思想,在标准最近邻域算法基础上提出了加权邻域数据关联算法(WNDA)。该算法综合考虑相关波门内的所有量测(包括正确量测和虚假量测)对状态的影响,提高了关联效果。同时算法不需要杂波密度等先验知识,不需要计算量测的关联概率,因而保持了较小的计算量。仿真结果表明,该方法有效地降低了误关联对跟踪效果的影响,同时保持了较小的计算量,在实际工程中有较好的应用前景。

Learning from correlation probability, this paper proposes a weighted neighbor data association algorithm （WNDA） based on the nearest neighbor standard filter （NNSF）. The method considers the impacts from all candidate measurements within correlation gate, no matter the measurements are true or false. The simulation results show that the proposed algorithm not only track well, but also keep small calculation cost, so, it can be applied in real work