基于量子遗传粒子滤波的无线传感器网络的目标跟踪算法的研究  

Research on Target Tracking Algorithm of Wireless Sensor Networks Based on Quantum Genetic Particle Filter

作　　者：魏颖[1] 郭鲁[1] Wei Ying;Guo Lu(School of Information and Control,Shenyang Institute of Technology,Fushun 113122,China)

机构地区：[1]沈阳工学院信息与控制学院,辽宁抚顺113122

出　　处：《计算机测量与控制》2021年第4期256-260,共5页Computer Measurement &Control

基　　金：国家自然科学基金(61603262,61403071);辽宁省自然科学基金(20180550418);沈阳工学院i5智能制造研究所基金(i5201701)。

摘　　要：文章主要研究的是关于量子遗传粒子滤波跟踪算法,该算法适用于无线传感器网络目标跟踪;用传统的粒子重采样可以解决粒子退化问题,但容易导致粒子逐渐减少,甚至耗尽;针对粒子耗尽问题,采用量子遗传算法,其独特的量子遗传的编码方式有效保证粒子组成的多样性,从而减缓了粒子滤波的退化现象,解决了粒子耗尽问题;通过引入量子的概念,同时让其并行可以有效地减少算法的运行时间,实时跟踪性能得到了大大提高;通过仿真结果表明该算法具有有效性、可行性;通过仿真:PF,GAPF、QGPF算法运行时间分别是56.16 s,46.71 s和30.46 s;全面比较后,QGPF算法计算时间最短;跟踪精度用均方根误差表示,与GAPF算法和PF算法相比QGPF算法位置和算法速度的中均方根误差最低(位置为0.0302,0.0258,速度0.0201,0.0101),其中PF算法的跟踪精度最低,QGPF跟踪精度最高,进一步表明QGPF算法具有良好的跟踪性能;实验结果表明,在大噪声条件下量子遗传粒子滤波算法改善了粒子贫乏问题,缩短跟踪时间和提高跟踪位置的精确性,鲁棒性,算法具有显著的优点。This paper mainly studies the quantum genetic particle filter tracking algorithm,which is suitable for target tracking in wireless sensor networks.With the gradual depletion of particles,the resampling problem can be solved easily.In order to solve the problem of particle depletion,quantum genetic algorithm(QGA)is used to solve the problem of particle depletion.The unique encoding method of quantum genetic algorithm can effectively ensure the diversity of particle composition,thus slowing down the degradation of particle filter and solving the problem of particle depletion.In this paper,by introducing the concept of quantum and making it parallel,the running time of the algorithm can be effectively reduced,and the real-time tracking performance is greatly improved.The simulation results show that the algorithm is effective and feasible.Through simulation:the running time of PF,GAPF and QGPF algorithm is 56.16 s,46.71 s and 30.46 s respectively.After a comprehensive comparison,the calculation time of QGPF algorithm is the shortest.Compared with GAPF algorithm and PF algorithm,QGPF algorithm has the lowest root mean square error of position and algorithm speed(position is 0.0302,0.0258,speed is 0.0201,0.0101),PF algorithm has the lowest tracking accuracy,and QGPF has the highest tracking accuracy.Further,it shows that QGPF algorithm has good tracking performance.The experimental results show that the quantum genetic particle filter algorithm can improve the particle impoverishment problem,shorten the tracking time and improve the accuracy and robustness of tracking position under the condition of large noise.

关 键 词：粒子滤波 目标跟踪 量子遗传算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]