摘要
针对资源有限的传感器网络中目标动态跟踪问题,提出了一种能够自适应选择跟踪传感器的机动目标协同跟踪算法。首先,采用粒子群优化算法优化传感器网络能耗与有效覆盖率,进行传感器位置部署;然后,以最大化候选传感器的Rényi信息增量与最小化传感器间信息传递能耗为适应度函数,采用二进制粒子群优化算法自适应选择最佳跟踪传感器组;最后,利用交互多模型粒子滤波对机动目标位置进行估计并进行分布式融合。仿真结果表明,与现有方法相比,该方法可在非高斯非线性环境下自适应选择最优跟踪传感器,显著提高目标跟踪精度,降低网络能耗。
Focusing on the dynamic tracking problem in resource constrained sensor networks, a new ma- neuvering target collaborative tracking algorithm with selecting tracking sensors adaptively is proposed. Firstly, the particle swarm optimization algorithm is us to trade-off the sensor networks＇ energy consumption and the ef- fective coverage of the target area, and obtain the optimized sensors location. Then, the tracking sensors are se- lected according to the maximal R6nyi information gain and the minimal energy consumption by a binary particle swarm optimization algorithm. Finally, the kinematic state of the maneuvering target is estimated by the inter- acting multiple model particle filtering algorithm, and the estimate states of the selected tracking sensors are fused. Simulation results show that the proposed algorithm can adaptively select tracking sensors, achieve the desired tracking accuracy and reduce network energy consumption compared with traditional methods in a non- linear non-Gaussian system.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2013年第2期304-309,共6页
Systems Engineering and Electronics
基金
长江学者和创新团队发展计划(IRT0645)
国家重大专项工程基础理论研究项目(G52809220262)
国家部委预研基金(9140A07030211DZ01)
中央高校基本科研业务费专项资金(K5051202036)资助课题
关键词
传感器网络
协同跟踪
信息论
粒子群优化
交互多模型粒子滤波
sensor network
collaborative tracking
information theory
particle swarm optimization
inter-acting multiple model particle filter
