《厦门大学学报（自然科学版）》

COP-SLAM(closed-form online pose-chain simultaneous locatization and mapping)是一种轻量型图优化算法,可以实时优化.但当局内特征点集中位于图像的特定区域时,其信息矩阵不能很好地表征里程计精度,影响实时优化效果.基于局内特征点的分布,引入局内点集的面积表示局内点的分布,提出了一种视觉里程计的后端图优化方法.实验结果表明,该优化方法有效地降低了相机绝对轨迹误差.

Closed-form online pose-chain simultaneous locatization and mapping(COP-SLAM)is a lightweight graph optimization algorithm that meets real-time constraints.However,it cannot be a good measure of visual odometry accuracy when inlier points are mainly distributed on the specific region of the image,affecting the real-time optimization results.A general approach to back-end graph optimization of visual odometry is proposed based on inlier distribution.The area of inlier set is introduced to represent inlier distribution.Experimental results have demonstrated that the method can effectively reduce absolute trajectory errors.

引言
1 COP-SLAM[7]
2 基于局内点分布的图优化方法
3 实验结果与分析
4 结论

图1 COP-SLAM的优化示意图<br/>Fig.1 Optimization diagram of COP-SLAM

图1 COP-SLAM的优化示意图
Fig.1 Optimization diagram of COP-SLAM

图2 KITTI00匹配特征点<br/>Fig.2 Matched feature points of KITTI00

图2 KITTI00匹配特征点
Fig.2 Matched feature points of KITTI00

表1 数据集的详细信息<br/>Tab.1 Details about datasets

表1 数据集的详细信息
Tab.1 Details about datasets

图3 实验框架<br/>Fig.3 Experimental framework

图3 实验框架
Fig.3 Experimental framework

图4 本研究优化结果<br/>Fig.4 Optimization result of this research

图4 本研究优化结果
Fig.4 Optimization result of this research

表2 KITTI的优化时间<br/>Tab.2 Optimize time of KITTI

表2 KITTI的优化时间
Tab.2 Optimize time of KITTI

表3 KITTI的绝对轨迹误差<br/>Tab.3 Absolute trajectory error of KITTI

表3 KITTI的绝对轨迹误差
Tab.3 Absolute trajectory error of KITTI

[1] CADENA C,CARLONE L,CARRILLO H,et al.Past,present,and future of simultaneous localization and mapping:towards the robust-perception age[J].IEEE Transactions on Robotics,2016,32(6):1309-1332.
[2] DELLAERT F,KAESS M.Square root SAM:simultaneous localization and mapping via square root information smoothing[J].International Journal of Robotics Research,2006,25(12):1181-1203.
[3] KAESS M,RANGANATHAN A,DELLAERT F.ISAM:incremental smoothing and mapping[J].IEEE Transactions on Robotics,2008,24(6):1365-1378.
[4] KONOLIGE K,GRISETTI G,KUMMERLE R,et al.Efficient sparse pose adjustment for 2D mapping[C]∥IEEE International Conference on Intelligent Robots and Systems.Taiwan:IEEE,2010:22-29.
[5] KUMMERLE R,GRISETTI G,STRASDAT H,et al.G2O:a general framework for graph optimization[C]∥IEEE International Conference on Robotics and Automation.Piscataway:IEEE,2011:3607-3613.
[6] DUBBELMAN G,BROWNING B.Closed-form online pose-chain SLAM[C]∥IEEE International Conference on Robotics and Automation.Karlsruhe:IEEE,2013:5190-5197.
[7] DUBBELMAN G,BROWNING B.COP-SLAM:closed-form online pose-chain optimization for visual SLAM[J].IEEE Transactions on Robotics,2015,31(5):1194-1213.
[8] SUNDERHAUF N.Robust optimization for simultaneous localization and mapping[D].Chemnitz:Chemnitz University of Technology,2012:13-193.
[9] THRUN S,LIU Y,KOLLER D,et al.Simultaneous localization and mapping with sparse extended information filters[J].The International Journal of Robotics Research,2004,23(7/8):693-716.
[10] GEIGER A,LENZ P,STILLER C,et al.Vision meets robotics:the KITTI dataset[J].International Journal of Robotics Research,2013,32(11):1231-1237.
[11] GEIGER A,ZIEGLER J,STILLER C.StereoScan:dense 3d reconstruction in real-time[C]∥IEEE Intelligent Vehicles Symposium.Baden-Baden:IEEE,2011:963-968.
[12] HAST A,NYSJÖ J,MARCHETTI A.Optimal RANSAC:towards a repeatable algorithm for finding the optimal set[J].Journal of Wscg,2013,21(1):21-30.
[13] BARBER C B,DOBKIN D P,HUHDANPAA H.The quickhull algorithm for convex hulls[J].Acm Transactions on Mathematical Software,1998,22(4):469-483.
[14] LABBE M,MICHAUD F.Appearance-based loop closure detection for online large-scale and long-term operation[J].IEEE Transactions on Robotics,2013,29(3):734-745.
[15] STURM J,ENGELHARD N,ENDRES F,et al.A benchmark for the evaluation of RGB-D SLAM systems[C]∥IEEE International Conference on Intelligent Robots and Systems.Algarve:IEEE,2012:573-580.
[16] ENGEL J,STÜCKLER J,CREMERS D.Large-scale direct SLAM with stereo cameras[C]∥IEEE International Conference on Intelligent Robots and Systems.Hamburg:IEEE,2015:1935-1942.
[17] FROST D P,KÄHLER O,MURRAY D W.Object-aware bundle adjustment for correcting monocular scale drift[C]∥IEEE International Conference on Robotics and Automation.Stockholm:IEEE,2016:4770-4776.

备注

引言

1 COP-SLAM[7]

2 基于局内点分布的图优化方法

3 实验结果与分析

4 结论

学报简介

备注

引言

1 COP-SLAM[7]

2 基于局内点分布的图优化方法

3 实验结果与分析

4 结 论

学报简介

4 结论