摘要
为解决常规的基于粒子滤波的目标跟踪算法,使用状态转移分布作为采样粒子的建议分布函数,没有考虑当前的观测值,从而造成定位时间长、定位精度低的问题.采用将最新的观测值融合到采样过程中,利用粒子群优化方法实现目标的跟踪,使得采样粒子集往后验概率密度分布较大的区域移动,从而显著地降低了精确定位所需的粒子数.研究结果表明:改进后的目标跟踪算法性能更优.
The conventional particle filter based target tracking algorithm uses a state transition distribution as the proposal distribution function of sampling particles, and it does not take the current observation values into account, which results in the long time and the low accuracy for positioning. To solve this problem, this paper incorporates the newest observations into the sampling process and uses particle swarm optimization, and the target tracking is completed. Through particle swarm optimization with this method, particles are moved to the regions where they have larger values of posterior density function. As a result, the impoverishment of the particle filter is over come and the number of particles needed for accurate location is reduced dramatically. Simulation experiments show that the improved target tracking algorithm has better performance.
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2016年第9期978-982,共5页
Journal of Liaoning Technical University (Natural Science)
关键词
粒子滤波
粒子群优化
目标跟踪
无线传感器网络
贝叶斯估计
particle filter
particle swarm optimization
target tracking algorithm
wireless sensor network
Bayesestimation
