摘要
针对传统射频识别(RFID)技术不适合移动机器人跟踪动态目标位姿,利用粒子滤波器融合RFID和激光雷达扫描数据,实现移动机器人对动态目标的位姿跟踪。首先利用RFID的相位差信息估计动态目标相位差速率,然后与激光雷达扫描动态目标估计的时间差速率进行匹配;最后选取最近邻的K个匹配点进行粒子滤波,从而实现动态目标的定位。最后,在机器人SCITOS G5平台上开展验证测试,实验结果表明,所提方法可以很好地定位动态目标,有效地减少定位误差。
The traditional radio frequency identification( RFID) technology is not suitable for localizing moving object using a mobile robot. In this paper,the moving object tracking of a mobile robot is achieved by fusing RFID and laser scanning with a particle filter.The phase-based velocity is firstlyestimated by the RFID phase differenceand then matched with the time-based velocity estimated by a 2D laser range finder. Finally,the K most matching points are selected to update the particle filter and the localization of a moving object is realized. The experiment is carried out on a SCITOS G5 robot. The results show that the proposed method can localize the moving target well and reduce the positioning error effectively.
作者
何静
刘冉
肖宇峰
梁高丽
付余路
He Jing;Liu Ran;Xiao Yufeng;Liang Gaoli;Fu Yulu(Robot Technology Used for Special Environment Key Laboratory of Sichuan Province, School of Information Engineering, Southwest University of Science and Technology,Mianyang 621010, China)
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2018年第2期81-88,共8页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(61601381,61471306,61701421)
国家核能开发科研项目([2016]1295)
四川省科研支撑计划(2015GZ0035)项目资助
关键词
射频识别
相位差
速度匹配
粒子滤波
radio frequency identification(RFID)
phase difference
velocity matching
particle filter