【摘要】 提出了一种基于射频识别(RFID)与激光传感器径向速度匹配的多目标实时识别与定位方法。将激光信息进行聚类处理,并估算出运动物体径向速度,然后与RFID相位差估算出的目标径向速度进行匹配,以固定大小的时间序列作为"时间窗口",在每个窗口内为每个RFID标签找到相似度最高的聚类,从而实现多个动态目标的识别和定位。实验在SCITOS G5机器人上进行了验证。结果表明:该方法的识别准确率可达94. 8%,定位精度可达0. 36 m。更多还原

【Abstract】 A multi-objects identification and localization method based on similarity matching of radial velocity estimated by laser and radio frequency identification( RFID) sensors is proposed. The information acquired by laser range finder is segmented into different clusters,and the radial velocities of moving objects are estimated based on the clustering information of adjacent time,these velocities are matched against the radial velocities computed from the RFID phase difference. Besides,to realize the identification and localization of multiple dynamic objects,this method uses a fixed-time series as"time window "to find the cluster with the highest similarity in each window. The experiment is validated on SCITOS G5 robot. The results show that the identification rate can reach94. 8 % and the localization accuracy can reach 0. 36 m.更多还原