基于地平面的单目视觉辅助激光雷达SLAM研究

Sequence	Proposed algorithm	LeGO-LOAM-noIMU	LeGO-LOAM-IMU
1	9.7763	72.3448	5.4013
2	12.9146	89.8510	37.6222
3	9.7763	86.6547	5.3066
4	9.7762	102.4516	6.2184
5	10.9626	86.0318	5.2739
6	9.7762	75.7008	6.0007
7	8.7765	67.9349	11.4841
8	12.9145	84.7132	6.7188
9	9.8208	92.2212	5.2530
10	12.9146	60.0927	7.7874
Total average	10.7147	81.9316	6.7739

查看原文

表 2不同算法运行时间统计

Table2. Runtime statistics of different algorithmsunit: ms

Module	Proposed algorithm	LeGO-LOAM-noIMU	LeGO-LOAM-IMU
Ego-motion undistortion	32.6	0	2.2
Ground removal and segmentation	21.7	22.1	21.6
Feature extract	4.8	6.2	4.8
Odometry	3.3	15.1	7.6
Total	62.4	43.4	36.2

查看原文

晏小彬, 彭道刚, 戚尔江. 基于地平面的单目视觉辅助激光雷达SLAM研究[J]. 光学学报, 2020, 40(24): 2428002. Xiaobin Yan, Daogang Peng, Erjiang Qi. Research on Ground-Plane-Based Monocular Aided LiDAR SLAM[J]. Acta Optica Sinica, 2020, 40(24): 2428002.

基于地平面的单目视觉辅助激光雷达SLAM研究下载： 1649次

图 1. 本文算法流程图

Fig. 1. Flow chart of the proposed algorithm

图 2. 通过激光点云和兴趣区域提取地平面特征点

Fig. 2. Extract ground features point from LiDAR cloud and ROI

图 3. 交比不变性校验后的匹配特征

Fig. 3. Feature correspondences after CRI verify

图 4. 激光里程计搜索特征点匹配过程

Fig. 4. Process of feature point correspondence searching in Lidar odometry

图 5. 特征点匹配过程对比。(a)直接搜索匹配;(b)直接搜索过程的局部细节;(c)经相机运动先验变换后搜索匹配;(d)变换后搜索过程的局部细节

Fig. 5. Comparison of feature point correspondence process. (a) Search the correspondence directly; (b) local details of direct search process; (c) search correspondence after transformed by camera motion a priori; (d) local details of search process after transform

图 6. 轨迹对比及真实场景图。(a)本文算法、LeGO-LOAM-IMU、LeGO-LOAM-noIMU估计轨迹与真实轨迹对比;(b)轨迹所对应位置的卫星地图

Fig. 6. Comparison of trajectory and the real scene. (a) Comparison of real trajectory and trajectories estimated by LeGO-LOAM-IMU, LeGO-LOAM-noIMU, and proposed, respectively; (b) satellite map of the trajectory

图 7. x,y,z轴方向上本文算法、LeGO-LOAM-IMU、LeGO-LOAM-noIMU估计值与真实值对比

Fig. 7. Comparison of the estimated value from proposed algorithm, LeGO-LOAM-IMU, LeGO-LOAM-noIMU and the real value on x,y, and z axis

图 8. 机器人及室内实验环境。(a)机器人;(b)实验室环境

Fig. 8. Robot and laboratory environment. (a) Robot; (b) laboratory environment

图 9. 实验室环境实验的估计轨迹

Fig. 9. Estimated trajectory of laboratory environment

图 10. 室外环境实验结果。(a)算法估计轨迹对比;(b)点云地图;(c)卫星地图

Fig. 10. Experimental results of outdoor environment. (a) Comparison of trajectory estimated by algorithms; (b) point cloud map; (c) satellite map

表 1不同算法绝对轨迹误差比较

Table1. Comparison of ATE of different algorithmsunit: m

表 2不同算法运行时间统计

Table2. Runtime statistics of different algorithmsunit: ms

关于本站 Cookie 的使用提示

全站搜索

基于地平面的单目视觉辅助激光雷达SLAM研究 下载： 1649次

图 1. 本文算法流程图

Fig. 1. Flow chart of the proposed algorithm

图 2. 通过激光点云和兴趣区域提取地平面特征点

Fig. 2. Extract ground features point from LiDAR cloud and ROI

图 3. 交比不变性校验后的匹配特征

Fig. 3. Feature correspondences after CRI verify

图 4. 激光里程计搜索特征点匹配过程

Fig. 4. Process of feature point correspondence searching in Lidar odometry

图 5. 特征点匹配过程对比。(a)直接搜索匹配;(b)直接搜索过程的局部细节;(c)经相机运动先验变换后搜索匹配;(d)变换后搜索过程的局部细节

Fig. 5. Comparison of feature point correspondence process. (a) Search the correspondence directly; (b) local details of direct search process; (c) search correspondence after transformed by camera motion a priori; (d) local details of search process after transform

图 6. 轨迹对比及真实场景图。(a)本文算法、LeGO-LOAM-IMU、LeGO-LOAM-noIMU估计轨迹与真实轨迹对比;(b)轨迹所对应位置的卫星地图

Fig. 6. Comparison of trajectory and the real scene. (a) Comparison of real trajectory and trajectories estimated by LeGO-LOAM-IMU, LeGO-LOAM-noIMU, and proposed, respectively; (b) satellite map of the trajectory

图 7. x,y,z轴方向上本文算法、LeGO-LOAM-IMU、LeGO-LOAM-noIMU估计值与真实值对比

Fig. 7. Comparison of the estimated value from proposed algorithm, LeGO-LOAM-IMU, LeGO-LOAM-noIMU and the real value on x,y, and z axis

图 8. 机器人及室内实验环境。(a)机器人;(b)实验室环境

Fig. 8. Robot and laboratory environment. (a) Robot; (b) laboratory environment

图 9. 实验室环境实验的估计轨迹

Fig. 9. Estimated trajectory of laboratory environment

图 10. 室外环境实验结果。(a)算法估计轨迹对比;(b)点云地图;(c)卫星地图

Fig. 10. Experimental results of outdoor environment. (a) Comparison of trajectory estimated by algorithms; (b) point cloud map; (c) satellite map

表 1不同算法绝对轨迹误差比较

Table1. Comparison of ATE of different algorithmsunit: m

表 2不同算法运行时间统计

Table2. Runtime statistics of different algorithmsunit: ms

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索

基于地平面的单目视觉辅助激光雷达SLAM研究下载： 1649次