[1]孙 弋,张笑笑.结合退火优化和遗传重采样的RBPF算法[J].西安科技大学学报,2020,(02):349-355.
 SUN Yi,ZHANG Xiao-xiao.RBPF algorithm based on annealing optimization and genetic resampling[J].Journal of Xi'an University of Science and Technology,2020,(02):349-355.
点击复制

结合退火优化和遗传重采样的RBPF算法(/HTML)
分享到:

西安科技大学学报[ISSN:1672-9315/CN:61-1434/N]

卷:
期数:
2020年02期
页码:
349-355
栏目:
出版日期:
2020-03-30

文章信息/Info

Title:
RBPF algorithm based on annealing optimization and genetic resampling
文章编号:
1672-9315(2020)02-0349-07
作者:
孙 弋张笑笑
(西安科技大学 通信与信息工程学院,陕西 西安 710054)
Author(s):
SUN YiZHANG Xiao-xiao
(College of Communication and Information Engineering,Xi'an University of Science and Technology,Xi'an 710054,China)
关键词:
粒子滤波 RBPF算法 提议分布 重采样 交叉变异
Keywords:
particle filter RBPF algorithm proposed distribution resampling cross mutation
分类号:
TP 242
文献标志码:
A
摘要:
RBPF是一种有效解决同时定位和建图的算法。传统的RBPF算法使用的粒子数目多并且频繁地执行重采样,导致粒子退化且估计能力下降,从而构建的栅格地图精度不高。针对上述缺点,对RBPF提出优化,首先将机器人的运动模型与观测模型结合作为其混合提议分布,同时利用退火参数优化混合提议分布,调控两者在提议分布中的比例,使其更加精确; 其次在重采样过程中根据粒子的权值对其进行分类,对高权重以及低权重粒子引入自适应遗传算法变异交叉操作,减少了重采样次数,有效维持了粒子多样性。在MATLAB上进行仿真验证,同时结合了Kobuki运动底盘在机器人操作系统(ROS)上进行实际验证。实验结果表明,与传统的RBPF算法相比,算法能够使用更少的粒子精确估计出机器人的位姿及路标,能够建立精度更高的栅格地图,并且具有更低的均方根误差和计算时间。
Abstract:
RBPF is an algorithm that effectively solves simultaneous positioning and mapping.The traditional RBPF algorithm uses a large number of particles and performs resampling frequently,resulting in particle degradation and reduced estimation ability,so that the constructed raster map is not accurate.In view of the above shortcomings,the RBPF is optimized.Firstly,the motion model of the robot is combined with the observation model to be its mixed proposal distribution.At the same time,the annealing parameters are used to optimize the mixed proposal distribution,and the ratio of the two in the proposed distribution is adjusted to make it more accurate.In the process of resampling,the particles are classified according to the weight of the particles,and the adaptive genetic algorithm mutation crossover operation is introduced to the high weight and low weight particles,which reduces the number of resampling and effectively maintains the particle diversity.Simulation verification was performed on MATLAB,together with the Kobuki motion chassis to conduct actual verification on the robot operating system(ROS).The experimental results show that compared with the traditional RBPF algorithm,the proposed algorithm can accurately estimate the pose and road signs of the robot using fewer particles,and then establish a higher precision raster map with lower root mean square error and less calculating time.

参考文献/References:

[1] 陈 卓,苏卫华,安慰宁,等.移动机器人SLAM与路径规划在ROS框架下的实现[J].医疗卫生装备,2017,38(2):109-113. CHEN Zhuo,SU Wei-hua,AN Wei-ning,et al.SLAM and path planning of mobile robot in ROS framework[J].Chinese Medical Equipment Journal,2017,38(2):109-113. [2]Endres F,Jürgen Hess,Engelhard N,et al.An evaluation of the RGB-D SLAM system[C]//2012 IEEE International Conference on Robotics and Automation IEEE,2012:1691-1696. [3]Cheein,Toibero,Sciascio D,et al.Monte carlo uncertainty maps-based for mobile robot autonomous SLAM navigation[C]//IEEE International Conference on Industrial Technology IEEE,2010:1433-1438. [4]刘 畅.基于扩展卡尔曼滤波的同步定位与地图构建(SLAM)算法研究进展[J].装备制造技术,2017(12):41-43. LIU Chang.Research progress of synchronous positioning and map construction(SLAM)algorithm based on extended Kalman filter[J].Equipment Manufacturing Technology,2017(12):41-43. [5]Grisetti G,Tipaldi G D,Stachniss C,et al.Fast and accurate SLAM with Rao–Blackwellized particle filters[J].Robotics and Autonomous Systems,2007,55(1):30-38. [6]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. WANG Zhi-yuan,CHENG Lan,XIE Gang.An improved particle filter algorithm and its application in multipath estimation[J].Computer Engineering,2017,43(6):289-295. [7]张宏伟.一种约束扩展卡尔曼粒子滤波器[J].东莞理工学院学报,2018,25(5):10-16. ZHANG Hong-wei.Constrained extended Kalman particle filter[J].Journal of Dongguan Institute of Technology,2018,25(5):10-16. [8]Yuvapoositanon P.Fast computation of look-ahead rao blackwellised particle filter in SLAM[C]//2014 International Electrical Engineering Congress(iEECON).IEEE,2014:1-4. [9]De-Yun M O Y W X.Hybrid system monitoring and diagnosing based on particle filter algorithm[J].Acta Automatica Sinica,2003,29(5):641-648. [10]Bailey T,Durrant-Whyte H.Simultaneous localization and mapping(SLAM):Part II[J].IEEE Robotics & Automation Magazine,2006,13(3):108-117. [11]Agarwal S,Shree V,Chakravorty S.RFM-SLAM:Exploiting relative feature measurements to separate orientation and position estimation in SLAM[C]//2017 IEEE International Conference on Robotics and Automation(ICRA).IEEE,2017:6307-6314. [12]禹鑫燚,朱熠琛,詹益安,等.SLAM过程中的机器人位姿估计优化算法研究[J].高技术通讯,2018(8):712-718. YU Xin-yi,ZHU Yi-chen,ZHAN Yi-an,et al.Research on optimization algorithm of robot pose estimation in slam process[J].High Technology Letters,2018(8):712-718. [13]强 锋.移动机器人室内地图构建及定位方法的相关分析[J].文化创新比较研究,2018(35):161-162. QIANG Feng.Correlation analysis of interior map construction and location method of Mobile robot[J].Comparative Study of Cultural Innovation,2018(35):161-162. [14]庄 严,王 伟,王 珂,等.移动机器人基于激光测距和单目视觉的室内同时定位和地图构建[J].自动化学报,2005,31(6):113-121. ZHUANG Yan,WANG Wei,WANG Ke,et al.Indoor simultaneous positioning and map construction of mobile robot based on laser ranging and monocular vision[J].ACTA Automatica Sinica,2005,31(6):113-121. [15]王法胜,鲁明羽,赵清杰,等.粒子滤波算法[J].计算机学报,2014,37(8):1679-1694. WANG Fa-sheng,LU Ming-yu,ZHAO Qing-jie,et al.Particle filter algorithm[J].Chinese Journal of Computers,2014,37(8):1679-1694. [16]王卫华,陈卫东,席裕庚.基于不确定信息的移动机器人地图创建研究进展[J].机器人,2001(6):563-568. WANG Wei-hua,CHEN Wei-dong,XI Yu-geng.Research progress on mobile robot map creation based on uncertain information[J].Robot,2001(6):563-568. [17]Yuen D C K,Macdonald B A.A comparison between extended Kalman filtering and sequential Monte Carlo techniques for simultaneous localisation and map-building[C]//Proceedings of the 2002 Australasian Conference on Robotics and Automation.ARAA,Auckland,New Zealand,2002:111-116. [18]骆燕燕,陈 龙.融合视觉信息的激光定位与建图[J].工业控制计算机,2017(12):21-23. LUO Yan-yan,CHEN Long.Laser slam based on fusion of visual information[J].Industrial Control Computer,2017(12):21-23. [19]朱 磊,樊继壮,赵 杰,等.改进粒子滤波器的移动机器人同步定位与地图构建方法[J].重庆大学学报,2014,37(4):39-45. ZHU Lei,FAN Ji-zhuang,ZHAO Jie,et al.Mobile robot synchronous positioning and map construction method based on improved particle filter[J].Journal of Chongqing University,2014,37(4):39-45. [20]厉茂海,洪炳熔,罗荣华.用改进的Rao-Blackwellized粒子滤波器实现移动机器人同时定位和地图创建[J].吉林大学学报(工学版),2007,32(2):401-406. LI Mao-hai,HONG Bing-rong,LUO Rong-hua.Mobile robot simultaneous positioning and map creation with improved Rao-Blackwellized particle filter[J].Journal of Jilin University(Engineering Edition),2007,32(2):401-406. [21]余洪山,王耀南.基于粒子滤波器的移动机器人定位和地图创建研究进展[J].机器人,2007(3):281-289,29. YU Hong-shan,WANG Yao-nan.Research progress of mobile robot positioning and map creation based on particle filter[J].Robot,2007(3):281-289,29. [22]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策,2005(4):361-365,37. HU Shi-qiang,JING Zhong-liang.Survey of particle filter algorithms[J].Control and Decision,2005(4):361-365,37. [23]马 成,朱 奕,伞 冶.一种基于区间估计的粒子滤波算法[J].哈尔滨工业大学学报,2013(11):8-12. MA Cheng,ZHU Yi,SAN Ye.A particle filter based on interval estimation[J].Journal of Harbin Institute of Technology,2013(11):8-12. [24]张瑞成,何丹阳.基于模拟退火粒子群算法的最大风能跟踪方法[J].工业控制计算机,2018(10):40-41,49. ZHANG Rui-cheng,HE Dan-yang.Maximum wind energy tracking method based on simulated annealing particle swarm optimization[J].Industrial Control Computer,2018(10):40-41,49. [25]Doucet A,De Freitas N,Murphy K,et al.Rao-blackwellised particle filtering for dynamic Bayesian networks[C]//Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence,2000:176-183. [26]郑 兵,陈世利,刘 蓉.基于萤火虫算法优化的Gmapping研究[J].计算机工程,2018,44(9):22-27. ZHENG Bing,CHEN Shi-li,LIU Rong.Gmapping research based on firefly algorithm optimization[J].Computer Engineering,2018,44(9):22-27. [27]张 毅,郑潇峰,罗 元,等.基于高斯分布重采样的Rao-Blackwellized粒子滤波SLAM算法[J].控制与决策,2016,31(12):2299-2304. ZHANG Yi,ZHENG Xiao-feng,LUO Yuan,et al.An improved particle filter algorithm and its application in multipath estimation[J].Control and Decision,2016,31(12):2299-2304. [28]王田橙,蔡云飞,唐振民.基于区域粒子群优化和部分高斯重采样的SLAM方法[J].计算机工程,2017,43(11):310-316. WANG Tian-cheng,CAI Yun-fei,TANG Zhen-min.SLAM method based on regional particle swarm optimization and partial Gaussian resampling[J].Computer Engineering,2017,43(11):310-316.

相似文献/References:

[1]张威虎,郭明香,贺元恺,等.一种改进的蝴蝶算法优化粒子滤波算法[J].西安科技大学学报,2019,(01):119.[doi:10.13800/j.cnki.xakjdxxb.2019.0117 ]
 ZHANG Wei-hu,GUO Ming-xiang,HE Yuan-kai,et al.An improved butterfly algorithm optimizing particle filter algorithm[J].Journal of Xi'an University of Science and Technology,2019,(02):119.[doi:10.13800/j.cnki.xakjdxxb.2019.0117 ]

备注/Memo

备注/Memo:
收稿日期:2019-09-20 责任编辑:高 佳
通信作者:孙 弋(1972-),男,陕西西安人,博士,教授,E-mail:409298131@qq.com
更新日期/Last Update: 2020-03-30