摘要
粒子群优化(PSO)算法的主要特点是能快速得到问题的解,缺点是容易陷入局部最优。提出了一种利用最佳维变异技术和量子理论方法改进的PSO算法,并应用于目标跟踪传感器调度问题。目标的动力学模型为线性高斯模型,传感器观测值被高斯噪声污染并与目标状态线性相关。对于多传感器单目标跟踪的数学问题,引入提出的最佳维变异PSO算法,在整个时间轴上产生最小成本。仿真实验结果表明:提出的算法比已有的算法收敛速度更快,全局搜索能力更强,传感器调度效率更高。
The main characteristic of the particle swarm optimization （PSO） algorithm is that it can get the solution quickly, and its disadvantage is easy to fall into a local optimum in the process. An improved PSO algorithm is proposed using best dimension mutation technique and quantum theory. It is applied to the sensor scheduling problem for target tracking. The dynamics model of the target is linear Gaussian model, and the sensor measurements impaired by white Gaussian noise are linearly related to the state of the target. A numerical problem of single target tracking with muhiple sensors is studied using the proposed best dimension mutation particle swarm optimization（BDMPSO） algorithm to get the minimal cost on the whole timeline. The simulation results show that the proposed algorithm has much faster convergence, stronger global search ability and more scheduling efficient than the existing algorithms.
出处
《传感器与微系统》
CSCD
北大核心
2011年第12期145-148,152,共5页
Transducer and Microsystem Technologies
基金
中央高校基本科研业务费专项资金资助项目(JUSRP111A46)
国家自然科学基金资助项目(61170119)
关键词
传感器调度
粒子群优化
目标跟踪
sensor scheduling
particle swarm optimization（PSO）
target tracking
