融合异质传感器信息的机器人精准室内定位  

Precise Robot Indoor Localization Based on Information Fusion of Heterogeneous Sensors

作　　者：张彬[1] 王永雄[1,2] 邵翔 秦琪[1] 符小媚[1] ZHANG Bin;Wang Yong-xiong;SHAO Xiang;QIN Qi;FU Xiao-mei(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093;Shanghai Engineering Research Center of Assistive devices,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093 [2]上海康复器械工程技术研究中心,上海200093

出　　处：《控制工程》2018年第7期1335-1343,共9页Control Engineering of China

基　　金：国家自然科学基金(61673276,61374039,61403254)

摘　　要：采用融合超宽带（UWB）测距模块、电子罗盘和电机编码器等异质传感器信息的方法实现高精度的机器人室内定位。首先,根据电机编码器信息采用相对定位法推算机器人的位姿1,同时采用基于UWB测距模块和电子罗盘的绝对定位法获取机器人的位姿2,最后通过卡尔曼滤波算法融合两种位姿信息,修正相对定位法的累积误差和绝对定位法的瞬时误差,实现高精度的机器人室内定位。实验结果表明了该方法的有效性,融合后的位姿精度相比于位姿1和位姿2的精度有显著的提高,坐标精度达到90%,航向角的精度达到97%以上。In order to obtain the high-precision indoor position and pose for mobile robots, we used the methods of dissimilar sensors, which include the ultra-wide band（UWB） ranging module, electronic compass and motor encoder. Firstly, according to the information of motor encoder, the relative position method is used to estimate the position 1 of the robot, and the absolute position method based on UWB ranging module and electronic compass is used to acquire the pose 2 of the robot. Finally, the Kalman filter algorithm is used to fuse the two poses information, and reduce the cumulative error of the relative positioning method and the instantaneous error of the absolute positioning method, to achieve high-precision indoor positioning for mobile robots. Experimental results show the effectiveness of our method. There are many significant improvements for the precision of robot poses after fusion compared to the accuracy of pose 1 and pose 2. The accuracy of acquired robot coordinates after fusion reaches 90 %, while that of the heading angle reaches over 97 %.

关 键 词：移动机器人 室内定位 UWB 卡尔曼滤波 数据融合 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]