多新息理论改进优化粒子滤波研究
|
摘要
研究多新息理论改进优化粒子滤波问题,以提高粒子滤波精度。其中粒子滤波状态估计模型的改进和新息数的选择是难点题。首先,基于多新息理论,利用多新息的增益矩阵和新息向量的乘积,对标准粒子滤波的状态估计模型改进优化,提出基于多新息改进优化的粒子滤波MI-PF;然后,分析了MI-PF算法的时间复杂度,从理论上论述了多新息理论改进优化粒子滤波的可行性;最后,利用仿真的方法,研究了不同新息数对MI-PF滤波性能和效率的影响,并得出了在强非线性、非高斯系统下,MI-PF新息数的取值范围。实验结果表明:若粒子数相同,与其它改进的粒子滤波相比,MI-PF滤波精度更高;若滤波精度近似,MI-PF使用的粒子数最少效率最高。从而验证了多信息理论改进粒子滤波的可行性,提高了粒子滤波精度。
Firstly,based on multi innovation theory,the product of gain matrix and vector of multi innovation were used to improve and optimize the state estimate value of standard particle filter,and the improvement and optimization of MI-PF was proposed based on multi innovation theory. Then,the time complex of MI-PF was analyzed,which discussed the feasibility of the improvement and optimization particle filter by multi innovation theory. Finally,using the method of experimental simulation,the influence of the different number of innovation on filtering precision was researched,and the number range of innovation was obtained with the system of non-liner and non-Gauss. The experimental results show that compared with other improved particle filters,if the particle number is the same,the MI-PF filter has higher accuracy; and if the filtering accuracy is approximate,MI-PF uses the least number of particles and has the highest efficiency. And then,the feasibility of improving particle filter by multi innovation was verified,and the accuracy of particle filter was improved.
Firstly,based on multi innovation theory,the product of gain matrix and vector of multi innovation were used to improve and optimize the state estimate value of standard particle filter,and the improvement and optimization of MI-PF was proposed based on multi innovation theory. Then,the time complex of MI-PF was analyzed,which discussed the feasibility of the improvement and optimization particle filter by multi innovation theory. Finally,using the method of experimental simulation,the influence of the different number of innovation on filtering precision was researched,and the number range of innovation was obtained with the system of non-liner and non-Gauss. The experimental results show that compared with other improved particle filters,if the particle number is the same,the MI-PF filter has higher accuracy; and if the filtering accuracy is approximate,MI-PF uses the least number of particles and has the highest efficiency. And then,the feasibility of improving particle filter by multi innovation was verified,and the accuracy of particle filter was improved.
引文
[1] V Chan. Theory and Applications of Monte Carlo Simulations[M].Rijeka,Croatia:In Teach,2013.
[2] N J Gordon,D J Salmond,A F M Smith. Novel approach to nonlinear/non-Gaussian Bayesian state estimation[J]. IEE Proceedings F:Radar and Signal Processing,1993,140(2):107-113.
[3] J Carpenter,P Clifford,P Fearnhead. Improved particle filter for non-linear problems[J]. IEE Proceedings F,1999,146(1):2-7.
[4]田梦楚,等.萤火虫算法智能优化粒子滤波[J].自动化学报,2016,42(1):89-97.
[5] Mai Thanh Nhat Truong,Sanghoon Kim. Parallel implementation of color-based particle filter for object tracking in embedded systems[J]. Human-centric Computing and Information Sciences,2017,7(1):1-13.
[6] K Madhan Kumar,A Velayudham,R Kanthavel. An Efficient Method for Road Tracking from Satellite Images Using Hybrid Multi-Kernel Partial Least Square Analysis and Particle Filter[J]. Journal of Circuits,Systems and Computers,2017,26(11):1750281.
[7] Zhao Zhiqiang,et al. Remarkable local resampling based on particle filter for visual tracking[J]. Multimedia Tools and Applications,2017,76(1):835-860.
[8]许伦辉,丛晓野.改进粒子滤波对人物跟踪的应用[J].计算机仿真,2014,31(1):344-347,387.
[9]王爱侠,赵越.基于改进粒子滤波算法的视频超分辨率重建[J].计算机工程,2015,41(4):263-266,272.
[10] S Sarkka. Bayesian Filtering and Smoothing[M]. Cambridge:Cambridge University Press,2013.
[11]杜正聪,辛强,邓寻.基于权值优化的粒子滤波算法研究[J].重庆师范大学学报(自然科学报),2015,32(3):124-129.
[12]孙海洋,张利.无人机跟踪场景下的粒子滤波算法的改进[J].计算机仿真,2017,34(2):84-87.
[13] J Zhong,Y Fung,M Dai. A biologically inspired improvement strategy for particle filter:Ant colony optimization assisted particle filter[J]. International Journal of Control,Automation,and Systems,2010,8(3):519-526.
[14] J Zhong,Y Fung. Case study and proofs of and colony optimization improved particle filter algorithm[J]. IET Control and Applications,2012,6(5):689-697.
[15] T Li,S Sun,T P Sattar,J M Corchado. Fight sample degeneracy and impoverishment in particle filters:A review of intelligent approaches[J]. Expert Systems with Applications,2014,41(8):3944-3954.
[16] T Li,S Sun,T Sattar. Adapting sample size in particle filters through KLD-resampling. Electronics Letters,2013,49(12):740-742.
[17] O Straka,M Simandl. Particle filter with adaptive samples size.Kybernetika,2011,47(3):385-400.
[18] F Ding,D Xiao and Ding Tao. Multi-innovation stochastic gradient identification methods[J]. Control Theory and Application,2003,20(6):870-874.
[19] F Ding. Several multi-innovation identification methods[J]. Digital Signal Processing,2010,20(4):1027-1039.
[20] F Ding. System identification. Part F:multi-innovation identification theory and methods[J]. Journal of Nanjing University of Information Science&Technology,2012,4(1):1-28.
[21]王法胜,等.粒子滤波算法[J].计算机学报,2014,37(8):1679-1694.
[22]王志远,程兰,谢刚.一种改进粒子滤波算法及其在多径估计中的应用[J].计算机工程,2017,43(6):289-295.
[2] N J Gordon,D J Salmond,A F M Smith. Novel approach to nonlinear/non-Gaussian Bayesian state estimation[J]. IEE Proceedings F:Radar and Signal Processing,1993,140(2):107-113.
[3] J Carpenter,P Clifford,P Fearnhead. Improved particle filter for non-linear problems[J]. IEE Proceedings F,1999,146(1):2-7.
[4]田梦楚,等.萤火虫算法智能优化粒子滤波[J].自动化学报,2016,42(1):89-97.
[5] Mai Thanh Nhat Truong,Sanghoon Kim. Parallel implementation of color-based particle filter for object tracking in embedded systems[J]. Human-centric Computing and Information Sciences,2017,7(1):1-13.
[6] K Madhan Kumar,A Velayudham,R Kanthavel. An Efficient Method for Road Tracking from Satellite Images Using Hybrid Multi-Kernel Partial Least Square Analysis and Particle Filter[J]. Journal of Circuits,Systems and Computers,2017,26(11):1750281.
[7] Zhao Zhiqiang,et al. Remarkable local resampling based on particle filter for visual tracking[J]. Multimedia Tools and Applications,2017,76(1):835-860.
[8]许伦辉,丛晓野.改进粒子滤波对人物跟踪的应用[J].计算机仿真,2014,31(1):344-347,387.
[9]王爱侠,赵越.基于改进粒子滤波算法的视频超分辨率重建[J].计算机工程,2015,41(4):263-266,272.
[10] S Sarkka. Bayesian Filtering and Smoothing[M]. Cambridge:Cambridge University Press,2013.
[11]杜正聪,辛强,邓寻.基于权值优化的粒子滤波算法研究[J].重庆师范大学学报(自然科学报),2015,32(3):124-129.
[12]孙海洋,张利.无人机跟踪场景下的粒子滤波算法的改进[J].计算机仿真,2017,34(2):84-87.
[13] J Zhong,Y Fung,M Dai. A biologically inspired improvement strategy for particle filter:Ant colony optimization assisted particle filter[J]. International Journal of Control,Automation,and Systems,2010,8(3):519-526.
[14] J Zhong,Y Fung. Case study and proofs of and colony optimization improved particle filter algorithm[J]. IET Control and Applications,2012,6(5):689-697.
[15] T Li,S Sun,T P Sattar,J M Corchado. Fight sample degeneracy and impoverishment in particle filters:A review of intelligent approaches[J]. Expert Systems with Applications,2014,41(8):3944-3954.
[16] T Li,S Sun,T Sattar. Adapting sample size in particle filters through KLD-resampling. Electronics Letters,2013,49(12):740-742.
[17] O Straka,M Simandl. Particle filter with adaptive samples size.Kybernetika,2011,47(3):385-400.
[18] F Ding,D Xiao and Ding Tao. Multi-innovation stochastic gradient identification methods[J]. Control Theory and Application,2003,20(6):870-874.
[19] F Ding. Several multi-innovation identification methods[J]. Digital Signal Processing,2010,20(4):1027-1039.
[20] F Ding. System identification. Part F:multi-innovation identification theory and methods[J]. Journal of Nanjing University of Information Science&Technology,2012,4(1):1-28.
[21]王法胜,等.粒子滤波算法[J].计算机学报,2014,37(8):1679-1694.
[22]王志远,程兰,谢刚.一种改进粒子滤波算法及其在多径估计中的应用[J].计算机工程,2017,43(6):289-295.