基于xtion和惯导融合的机器人定位研究  

Research on Robot Positioning Based on Xtion and Inertial Fusion

作　　者：刘振宇[1] 张德喜 田大吉 LIU Zhen-yu;ZHANG De-xi;TIAN Da-ji(Shenyang University of Technology,Shenyang Liaoning 110870,China;SIASUN Robot&Automation CO.,LTD,ShenyangLiaoning 110168,China)

机构地区：[1]沈阳工业大学,辽宁沈阳110870 [2]新松机器人自动化股份有限公司,辽宁沈阳110168

出　　处：《计算机仿真》2020年第5期291-295,共5页Computer Simulation

摘　　要：为改善移动机器人基于视觉定位时在拐角及空旷地带跟踪丢失和精度不高等问题,提出了xtion深度相机和惯导信息融合的室内定位方法。首先使用ORB-SLAM2算法来计算移动机器人在相邻图像间的位姿。然后采用回环检测方法把位姿信息提供给后端进行优化,得到全局一致的轨迹。接着对惯导进行解算得到位姿信息,最后利用扩展卡尔曼滤波融合xtion深度相机和惯导各自的位姿信息来改善定位精度和稳定性。实验结果表明,上述方法能够提高室内移动机器人的定位精度和稳定性。In order to improve the tracking loss and accuracy of mobile robots based on visual positioning in corners and open areas, an indoor positioning method based on Xtion depth camera and inertial navigation information fusion is proposed. Firstly, the ORB-SLAM2 algorithm was used to calculate the pose of the mobile robot between adjacent images. Then, the pose detection information was provided to the back end for optimization by the loop detection method to obtain a globally consistent trajectory.Next, the INS was solved to obtain the pose information. Finally, the extended Kalman filter was used to fuse the positional information of the Xtion depth camera and the inertial navigation to improve the positioning accuracy and stability. The experimental results show that the method can improve the positioning accuracy and stability of indoor mobile robots.

关 键 词：定位 深度相机 惯导 扩展卡尔曼滤波 信息融合 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]