万有引力优化的粒子滤波算法

doi:10.3969/j.issn.1001-2400.2018.02.024

摘要/Abstract

摘要：

针对经典粒子滤波中存在的粒子易退化、易丧失多样性以及滤波精度严重依赖于粒子数量的问题，提出一种万有引力优化的粒子滤波算法．通过万有引力算法优化粒子滤波中的粒子集来提高滤波精度．首先将每个粒子看做质量大小正比于粒子权值的点，粒子间的引力吸引着粒子向高似然区域移动，从而优化粒子集．然后利用精英粒子策略加快万有引力优化算法中粒子收敛速度，并避免粒子陷入局部最优; 引入感知模型防止过度收敛导致的粒子拥挤或重叠．仿真实验表明，该算法在粒子数较少的情况下与经典粒子滤波算法和粒子群优化粒子滤波算法相比，保持了更好的粒子滤波精度和速度．

关键词: 粒子滤波, 粒子退化, 粒子贫化, 万有引力, 状态估计

Abstract:

As the traditional particle filter has problems of particle degeneracy and particle diversity loss and filter accuracy depends heavily on the particle number, a gravity optimized particle filter algorithm is proposed.The particle swarm is optimized by the gravity algorithm in the particle filter to improve the filtering accuracy. Each particle is regarded as a mass point and the mass is proportional to the particle weight.The gravity attracts particles moving toward the high likelihood region which optimizes the particle swarm.Then elite particle strategy is introduced to accelerate the particle convergence rate and avoid the local optimum in the gravity algorithm. The perceptual model is used to prevent particles from crowding or overlapping due to excessive convergence. Simulation results show that the proposed algorithm has a better filtering accuracy and speed in the case of few particles compared with the classical particle filter algorithm and particle swarm optimization particle filter algorithm.

Key words: particle filter, particle degeneracy, particle impoverishment, gravitation, state estimation

刘润邦;朱志宇. 万有引力优化的粒子滤波算法[J]. 西安电子科技大学学报, 2018, 45(2): 141-147.

LIU Runbang;ZHU Zhiyu. Gravity optimized particle filter algorithm[J]. Journal of Xidian University, 2018, 45(2): 141-147.

参考文献

［1］王泽玉, 李明, 张鹏. 一种新的进化裂变粒子滤波算法［J］. 西安电子科技大学学报, 2014, 41(6): 31-36.
WANG Zeyu, LI Ming, ZHANG Peng. Novel Particle Filter Algorithm Based on Evolution Fission［J］. Journal of Xidian University, 2014, 41(6): 31-36.
［2］巨刚, 袁亮, 刘小月. 多方法融合的粒子滤波算法的神经丝自动跟踪［J］. 西安电子科技大学学报, 2016, 43(4): 184-190.
JU Gang, YUAN Liang, LIU Xiaoyue. Neurofilament Protein Automatic Tracking of the Particle Filter Algorithm Based on Multiple Methods Fusion［J］. Journal of Xidian University, 2016, 43(4): 184-190.
［3］别秀德, 刘洪彬, 常发亮, 等. 自适应分块的多特征融合多目标跟踪［J］. 西安电子科技大学学报, 2017, 44(2): 151-157.
BIE Xiude, LIU Hongbin, CHANG Faliang, et al. Multi-target Tracking Method Based on the Adaptive Fragment and Multi-feature Fusion［J］. Journal of Xidian University, 2017, 44(2): 151-157.
［4］陈志敏, 田梦楚, 吴盘龙, 等. 基于蝙蝠算法的粒子滤波法研究［J］. 物理学报, 2017, 66(5): 47-56.
CHEN Zhimin, TIAN Mengchu, WU Panlong, et al. Intelligent Particle Filter Based on Bat Algorithm［J］. Acta Physica Sinica, 2017, 66(5): 47-56.
［5］王尔申, 庞涛, 曲萍萍, 等. 基于混沌的改进粒子群优化粒子滤波算法［J］. 北京航空航天大学学报, 2016, 42(5): 885-890.
WANG Ershen, PANG Tao, QU Pingping, et al. Improved Particle Filter Algorithm Based on Chaos Particle Swarm Optimization［J］. Journal of Beijing University of Aeronautics and Astronautics, 2016, 42(5): 885-890.
［6］ ZHAO Z S, FENG X, LIN Y Y, et al. Improved Rao-blackwellized Particle Filter by Particle Swarm Optimization［J］. Journal of Applied Mathematics, 2013, 2013( 2) : 1-7.
［7］田梦楚, 薄煜明. 陈志敏, 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.
TIAN Mengchu, BO Yuming, CHEN Zhimin, et al. Firefly Algorithm Intelligence Optimized Particle Filter［J］. Acta Automatica Sinica, 2016, 42(1): 89-97.
［8］ GAO M L, LI L L, SUN X M, et al. Firefly Algorithm(FA) Based Particle Filter Method for Visual Tracking［J］. Optik, 2015, 126(18): 1705-1711.
［9］ WANG H, WANG W J, ZHOU X Y, et al. Firefly Algorithm with Neighborhood Attraction［J］. Information Sciences, 2017, 382/383: 374-387.
［10］刘繁明, 钱东, 刘超华. 一种人工物理优化的粒子滤波算法［J］. 控制与决策, 2012, 27(8): 1145-1148.
LIU Fanming, QIAN Dong, LIU Chaohua. An Artificial Physics Optimized Particle Filter［J］. Control and Decision, 2012, 27(8): 1145-1148.
［11］ MORSHIDI M, TJAHJADI T. Gravity Optimised Particle Filter for Hand Tracking［J］. Pattern Recognition, 2014, 47(1): 194-207.
［12］ DAS P K, BEHERA H S, PANIGRAHI B K. A Hybridization of an Improved Particle Swarm Optimization and Gravitational Search Algorithm for Multi-robot Path Planning［J］. Swarm and Evolutionary Computation, 2016, 28(1): 14-28.
［13］ ALI A F, TAWHID M A. Direct Gravitational Search Algorithm for Global Optimisation Problems［J］. East Asian Journal on Applied Mathematics, 2016, 6(3): 290-313.
［14］马力, 刘丽涛. 万有引力搜索算法的分析与改进［J］. 微电子学与计算机, 2015, 32(9): 76-80.
MA Li, LIU Litao. Analysis and Improve of Gravitational Search Algorithm［J］. Microelectronics & Computer, 2015, 32(9): 76-80.
［15］ SIDDIQUE N, ADELI H. Gravitational Search Algorithm and Its Variants［J］. International Journal of Pattern Recognition and Artificial Intelligence, 2016, 30(8): 1639001.
［16］ ZHANG A, SUN G, REN J, et al. A Dynamic Neighborhood Learning-based Gravitational Search Algorithm［J］. IEEE Transactions on Cybernetics, 2016, 99(1): 1-12.

[1]	赵季红;谢志勇;曲桦;王明欣;刘熙. 一种简化鲁棒的传感网络节点三维估计算法[J]. 西安电子科技大学学报, 2018, 45(5): 108-114.
[2]	贺磊南;卢虎;姬正洲. WiFi/PDR室内分层三维定位技术[J]. 西安电子科技大学学报, 2017, 44(4): 167-173+166.
[3]	别秀德;刘洪彬;常发亮;彭志勇. 自适应分块的多特征融合多目标跟踪[J]. 西安电子科技大学学报, 2017, 44(2): 151-157.
[4]	王海环;王俊. 一种改进的多伯努利多目标跟踪算法[J]. 西安电子科技大学学报, 2016, 43(6): 176-182.
[5]	巨刚;袁亮;刘小月. 多方法融合的粒子滤波算法的神经丝自动跟踪[J]. 西安电子科技大学学报, 2016, 43(4): 184-190.
[6]	覃兵;田军;查宇飞;张立朝;黄宏图. 利用二值描述符的实时目标跟踪算法[J]. J4, 2015, 42(5): 168-174.
[7]	王尔申;张淑芳;蔡明;庞涛. 遗传粒子滤波的GPS接收机自主完好性监测[J]. J4, 2015, 42(1): 136-141.
[8]	赵雪刚;宋骊平;姬红兵. 量化量测条件下的交互多模型箱粒子滤波[J]. J4, 2014, 41(6): 37-44.
[9]	王泽玉;李明;张鹏. 一种新的进化裂变粒子滤波算法[J]. J4, 2014, 41(6): 31-36+75.
[10]	张鹏;张林让;胡子军. HPRF雷达距离延拓检测前跟踪方法[J]. J4, 2014, 41(5): 207-212.
[11]	卢锦;苏洪涛;水鹏朗. 采用代价参考粒子滤波器估计天波雷达目标状态[J]. J4, 2013, 40(5): 20-25+59.
[12]	李浩;王润亮;彭华. 粒子滤波盲均衡译码联合算法[J]. J4, 2013, 40(1): 123-128+176.
[13]	刘钦;刘峥;刘俊. 一种多机动目标协同跟踪的博弈论算法[J]. J4, 2012, 39(6): 49-54+135.
[14]	刘学;焦淑红;蓝晓宇. 拟蒙特卡罗聚合重采样粒子滤波无源定位算法[J]. J4, 2012, 39(5): 154-160.
[15]	李远征;卢朝阳;李静. 一种基于多特征融合的视频目标跟踪方法[J]. J4, 2012, 39(4): 1-6+45.