信息与控制2012，Vol.41Issue(6)：760-766,773,8.DOI:10.3724/SP.J.1219.2012.00760
基于GRNN的拟蒙特卡洛粒子滤波目标跟踪算法
Quasi-Monte Carlo Particle Filter Algorithm for Target Tracking Based on GRNN
摘要
Abstract
Quasi-Monte-Carlo particle filter (QMC-PF) can not meet the requirement of target tracking because of the high computational complexity. A novel Quasi-Monte-Carlo particle filter (NQMC-PF) algorithm for maneuvering radar target tracking is proposed. The algorithm applies QMC algorithm to generating the low-discrepancy offsprings of the the particles with heavy weight to replace the particles with low weight, which guarantees the quality and diversity of samples. Generalized regression neural network (GRNN) is used to calculate the weights of the offsprings, which improves the precision and the speed of the filter. The simulation results show that the calculation precision of the algorithm is higher than standard QMC-PF, and it possesses the advantages of short computation time and real-time standard. It can be applied to the radar target tracking.关键词
粒子滤波/拟蒙特卡洛方法/广义回归神经网络(GRNN)/目标跟踪/闪烁噪声Key words
particle filter/ quasi-Monte-Carlo algorithm/ generalized regression neural network (GRNN): target tracking/ glint noise分类
信息技术与安全科学引用本文复制引用
陈志敏,薄煜明,吴盘龙,徐文康,刘正凡..基于GRNN的拟蒙特卡洛粒子滤波目标跟踪算法[J].信息与控制,2012,41(6):760-766,773,8.基金项目
国家自然科学基金资助项目(61104196) （61104196）
高等学校博士学科点专项科研基金资助项目 (20113219110027). （20113219110027）
