摘要
传感器本身及外界环境存在的随机干扰,使得量测信息具有不确定性及相关性,且当某个传感器的测量精度较低时,直接融合观测信息导致融合算法有效性的降低。针对此问题,提出一种滤波系数化的分布式航迹融合方法。该方法通过局部估计信息建立支持度矩阵来完成滤波估计信息的系数化,最后加权组合滤波,实现对目标的实时融合跟踪。相对于原有融合算法,无论在哪种情况下该方法均使融合跟踪精度提高了16%以上,保证了算法的有效性。
The multi-sensor interference and environment noises make observations uncertain and correlative,besides,when the measurement accuracy of a certain sensor is low,the validity of the fusion algorithm is reduced because of fusing the observations directly.For this problem,a distributed track fusion method of filtering coefficients is proposed.It uses local state estimates to establish the matrix of supporting degree to complete filtering coefficients.And then it combines all weighted filtering values to fulfill the second filter so that it achieves target tracking in real-time.Simulation results show that the method improves tracking accuracy further and ensures algorithmic validity.
出处
《计算机工程与应用》
CSCD
北大核心
2011年第36期141-143,179,共4页
Computer Engineering and Applications
基金
航空科学基金(No.20090580013)
中央高校基础研究基金(No.ZYGX2009J092)
关键词
多传感器
滤波系数化
测量精度
航迹融合
multi-sensor
filtering coefficients
measurement accuracy
track fusion
