| 摘 要: 为了提升室内轮式机器人在移动过程中的定位精度，提出一种基于多信息融合的定位算法。鉴于单一传感器的定位缺陷，使用多传感器信息进行数据融合来提升定位精度，将机器人里程计和惯性测量单元(Inertia Measurement Unit, IMU)解算的位姿信息通过扩展卡尔曼滤波进行信息融合，组成里程计融合器；然后将里程计融合器解算的位姿信息与激光雷达的点云信息进行融合，组成激光雷达融合器；最后根据不同的地图环境，对里程计融合器和激光雷达融合器的数据进行加权融合，组成加权融合器。经过实验验证，多传感器的融合相对于单一传感器平均定位精度提高了约60%。 |
| 关键词: 机器人定位；ICP；多信息融合；扩展卡尔曼滤波 |
|
中图分类号: TP391
文献标识码: A
|
|
| Research on Positioning Algorithm of Wheeled Robot based on Multi-information Fusion |
|
SHEN Nianwei, YU Dayong
|
(School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China)
17853589237@163.com; 825298801@qq.com
|
| Abstract: In order to improve the positioning accuracy of indoor wheeled robots in the moving process, this paper proposes a positioning algorithm based on multi-information fusion. In view of the positioning defect of a single sensor, multi-sensor information is used for data fusion to improve the positioning accuracy. The position and pose information calculated by robot odometer and IMU(Inertia Measurement Unit) is fused through extended Kalman filter to form an odometer fuser. Then, the position and pose information calculated by the odometer fuser is fused with the point cloud information of the lidar to form a lidar fuser. Finally, according to different map environments, the data of odometer fuser and lidar fuser are weighted and fused to form a weighted fuser. It has been verified by experiments that the average positioning accuracy of multi-sensor fusion improves by about 60%, compared to that of a single sensor. |
| Keywords: robot positioning; ICP(Iterative Closest Point); multi-information fusion; extended Kalman filter |
