Visual-aid inertial SLAM method based on graph optimization in indoor

被引：0

作者：

Xu X.-S. ^{[1
,2
]}

Dai W. ^{[1
,2
]}

Yang B. ^{[1
,2
]}

Li Y. ^{[1
,2
]}

Dong Y. ^{[1
,2
]}

机构：

[1] Key Laboratory of Micro-inertial Instrument and Advanced Navigation Technology, Ministry of Education, Nanjing

[2] School of Instrument Science and Engineering, Southeast University, Nanjing

来源：

| 2017年 / Editorial Department of Journal of Chinese Inertial Technology卷 / 25期

关键词：

Cluster sampling; Graph optimization; Integrated location; Loop detection; Simultaneous localization and mapping (SLAM);

D O I：

10.13695/j.cnki.12-1222/o3.2017.03.007

中图分类号：

学科分类号：

摘要：

To realize the accurate autonomous navigation of mobile robot in unfamiliar environment, a graph-based SLAM method based on the visual odometry is proposed, which adds loop detection into the motion estimation of visual odometry and thus optimizes the nonlinear constraints to improve the positioning accuracy. In the motion estimation of visual odometry, a cluster sampling matching track method based on ORB feature is be proposed to deal with the high mismatch rate. In the pose graph optimization, an improved loop detection method is proposed to reduce the mismatch possibility. Finally, the inertial navigation system is combined with the visual SLAM to further improve the steady and accuracy of the system. The simulations using open indoor SLAM test datasets show that the locating RMSE of the proposed method is in centimeter level, and the generated point cloud map is clearly visible. © 2017, Editorial Department of Journal of Chinese Inertial Technology. All right reserved.

引用

页码：313 / 319

页数：6

共 13 条

[11] Liu Y., Cormack L.K., Bovik A.C., Statistical modeling of 3-D natural scenes with application to Bayesian stereosis, IEEE Transactions on Image Processing, 20, 9, pp. 2515-2530, (2011)
[12] Skibinski S., Weichert F., Parametric fusion of complex landmark observations present within the road network by utilizing bundle-adjustment-based full-SLAM, pp. 917-926, (2016)
[13] Wang Y.-S., Liu J.-Y., Zeng Q.-H., Et al., 3D environment restructure method with structured light for indoor vision/inertial navigation, Journal of Chinese Inertial Technology, 24, 1, pp. 51-58, (2016)

← 1 2 →