基于鱼眼相机和RGBD相机的异构协同SLAM
收稿日期: 2022-06-14
修回日期: 2022-08-04
录用日期: 2022-09-15
网络出版日期: 2022-09-30
基金资助
国家自然科学基金(61873031)
Heterogeneous collaborative SLAM based on fisheye and RGBD cameras
Received date: 2022-06-14
Revised date: 2022-08-04
Accepted date: 2022-09-15
Online published: 2022-09-30
Supported by
National Natural Science Foundation of China(61873031)
SLAM技术在卫星导航拒止环境下的自主导航中有着广泛的应用前景。结合单目鱼眼SLAM可获取更多纹理信息的优势和RGBD-SLAM可直接获取尺度信息的优势,设计基于单目鱼眼相机和RGBD相机的异构协同SLAM系统。首先设计特征点对三维灰度质心方向一致性检验方法以筛选异构图像之间的候选匹配点。然后设计异构图像之间的分步式光流-投影匹配方法,以实现鱼眼相机和RGBD相机之间高性能的特征点匹配与相对位姿估计。最后基于ORB-SLAM2框架,提出基于鱼眼相机和RGBD相机的异构协同SLAM系统框架。实验结果表明:相比于传统的特征点匹配方法,设计的特征匹配方法在异构相机的图像特征匹配任务中能够表现出更高的性能。相比于单目鱼眼SLAM系统和RGBD-SLAM系统,提出的异构协同SLAM系统在相机快速移动、相机贴近景物、低帧率、纹理缺失等条件下,以及在相机纯旋转运动、室外大场景等条件下性能更优,其鲁棒性、抗轨迹漂移能力和轨迹精度比单目鱼眼SLAM系统和RGBD-SLAM系统都有较大提升。
张宇桐 , 宋建梅 , 丁艳 , 刘锦鹏 . 基于鱼眼相机和RGBD相机的异构协同SLAM[J]. 航空学报, 2023 , 44(10) : 327621 -327621 . DOI: 10.7527/S1000-6893.2022.27621
SLAM technology has a the potential of wide applications in autonomous navigation in satellite navigation denied environments. To combine the advantages of monocular fisheye SLAM to obtain more texture information and RGBD-SLAM to directly obtain scale information, a heterogeneous collaborative SLAM system is designed based on a monocular fisheye camera and a RGBD camera. Firstly, a method for checking the 3D gray centroid direction consistency between feature points is designed to screen the candidate matching points between heterogeneous images. Then, a step-by-step optical flow and projection matching method between heterogeneous images is designed to achieve high-performance feature point matching and relative pose estimation between fisheye and RGBD camera. Finally, based on the ORB-SLAM2 framework, a heterogeneous collaborative SLAM system is proposed based on fisheye and RGBD camera. The experimental results show that compared with traditional feature point matching methods, the proposed feature point matching method shows higher performance in the task of image feature matching with heterogeneous cameras. Compared with the monocular fisheye SLAM and RGBD-SLAM system, the proposed heterogeneous collaborative SLAM system has better performance under the conditions of rapid camera movement, camera close to the scene, low frame rate, texture loss, pure rotation of the camera, outdoor large scenes, etc., and demonstrates improved robustness, anti-drift ability and trajectory accuracy.
1 | 赵良玉, 金瑞, 朱叶青, 等. 基于点线特征融合的双目惯性SLAM算法[J]. 航空学报, 2022, 43(3): 325117. |
ZHAO L Y, JIN R, ZHU Y Q, et al. Stereo visual-inertial slam with point and line features[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(3): 325117 (in Chinese). | |
2 | 谢洪乐, 陈卫东, 范亚娴, 等. 月面特征稀疏环境下的视觉惯性SLAM方法[J]. 航空学报, 2021, 42(1): 524169. |
XIE H L, CHEN W D, FAN Y X, et al. Visual-inertial SLAM in featureless environments on lunar surface[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(1): 524169 (in Chinese). | |
3 | 王小涛, 张家友, 王邢波, 等. 基于FastSLAM的绳系机器人同时定位与地图构建算法[J]. 航空学报, 2021, 42(1): 523893. |
WANG X T, ZHANG J Y, WANG X B, et al. Simultaneous localization and mapping algorithm based on FastSLAM framework for tethered robots[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(1): 523893 (in Chinese). | |
4 | DAVISON A J, REID I D, MOLTON N D, et al. MonoSLAM: Real-time single camera SLAM[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(6): 1052-1067. |
5 | KLEIN G, MURRAY D. Parallel tracking and mapping for small AR workspaces[C]∥ 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality. Piscataway: IEEE Press, 2008: 225-234. |
6 | KLEIN G, MURRAY D. Improving the agility of keyframe-based SLAM[C]∥European Conference on Computer Vision. Berlin: Springer, 2008: 802-815. |
7 | MUR-ARTAL R, MONTIEL J M M, TARDóS J D. ORB-SLAM: A versatile and accurate monocular SLAM system[J]. IEEE Transactions on Robotics, 2015, 31(5): 1147-1163. |
8 | LIN Y, GAO F, QIN T, et al. Autonomous aerial navigation using monocular visual-inertial fusion[J]. Journal of Field Robotics, 2018, 35(1): 23-51. |
9 | WANG Y H, CAI S J, LI S J, et al. CubemapSLAM: A piecewise-pinhole monocular fisheye SLAM system[C]∥Asian Conference on Computer Vision. Cham: Springer, 2019: 34-49. |
10 | CARUSO D, ENGEL J, CREMERS D. Large-scale direct SLAM for omnidirectional cameras[C]∥2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Piscataway: IEEE Press, 2015: 141-148. |
11 | ENGEL J, SCH?PS T, CREMERS D. LSD-SLAM: Large-scale direct monocular SLAM[C]∥European Conference on Computer Vision. Cham: Springer, 2014: 834-849. |
12 | ZHANG Z C, REBECQ H, FORSTER C, et al. Benefit of large field-of-view cameras for visual odometry[C]∥ 2016 IEEE International Conference on Robotics and Automation (ICRA). Piscataway: IEEE Press, 2016: 801-808. |
13 | FORSTER C, PIZZOLI M, SCARAMUZZA D. SVO: Fast semi-direct monocular visual odometry[C]∥2014 IEEE International Conference on Robotics and Automation (ICRA). Piscataway: IEEE Press, 2014: 15-22. |
14 | MATSUKI H, VON STUMBERG L, USENKO V, et al. Omnidirectional DSO: Direct sparse odometry with fisheye cameras[J]. IEEE Robotics and Automation Letters, 2018, 3(4): 3693-3700. |
15 | ENGEL J, KOLTUN V, CREMERS D. Direct sparse odometry[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(3): 611-625. |
16 | JO Y G, HONG S H, HWANG S S, et al. Fisheye lens camera based autonomous valet parking system[DB/OL]. arXiv preprint: 2104.13119, 2021. |
17 | LIU S Y, GUO P, FENG L H, et al. Accurate and robust monocular SLAM with omnidirectional cameras[J]. Sensors (Basel, Switzerland), 2019, 19(20): 4494. |
18 | KHOMUTENKO B, GARCIA G, MARTINET P. An enhanced unified camera model[J]. IEEE Robotics and Automation Letters, 2016, 1(1): 137-144. |
19 | CORREIA GARCIA T A, CAMPOS M B, CASTANHEIRO L F, et al. A proposal to integrate ORB-slam fisheye and convolutional neural networks for outdoor terrestrial mobile mapping[C]∥ 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. Piscataway: IEEE Press, 2021: 578-581. |
20 | URBAN S, HINZ S. MultiCol-SLAM - A modular real-time multi-camera SLAM system[DB/OL]. arXiv preprint: 1610.07336, 2016. |
21 | JI S P, QIN Z J, SHAN J. Panoramic SLAM from a multiple fisheye camera rig[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 159: 169-183. |
22 | MUR-ARTAL R, TARDóS J D. ORB-SLAM2: An open-source SLAM system for monocular, stereo, and RGB-D cameras[J]. IEEE Transactions on Robotics, 2017, 33(5): 1255-1262. |
23 | ZHANG Y, HUANG F. Panoramic visual SLAM technology for spherical images[J]. Sensors (Basel, Switzerland), 2021, 21(3): 705. |
24 | ZHAO Q, FENG W, WAN L, et al. SPHORB: A fast and robust binary feature on the sphere[J]. International Journal of Computer Vision, 2015, 113(2): 143-159. |
25 | AUDRAS C, COMPORT A I, MEILLAND M, et al. Real-time dense appearance-based SLAM for RGB-D sensors[C]∥Proceedings of the 2011 Australasian Conference on Robotics and Automation, 2011. |
26 | NEWCOMBE R A, IZADI S, HILLIGES O, et al. KinectFusion: Real-time dense surface mapping and tracking[C]∥ 2011 10th IEEE International Symposium on Mixed and Augmented Reality. Piscataway: IEEE Press, 2012: 127-136. |
27 | WHELAN T, JOHANNSSON H, KAESS M, et al. Robust real-time visual odometry for dense RGB-D mapping[C]∥ 2013 IEEE International Conference on Robotics and Automation. Piscataway: IEEE Press, 2013: 5724-5731. |
28 | CHEN J W, BAUTEMBACH D, IZADI S. Scalable real-time volumetric surface reconstruction[J]. ACM Transactions on Graphics, 2013, 32(4): 1-16. |
29 | ENDRES F, HESS J, STURM J, et al. 3-D mapping with an RGB-D camera[J]. IEEE Transactions on Robotics, 2014, 30(1): 177-187. |
30 | HENRY P, KRAININ M, HERBST E, et al. RGB-D mapping: Using depth cameras for dense 3D modeling of indoor environments[M]∥Experimental Robotics. Berlin: Springer, 2014: 477-491. |
31 | BESL P J, MCKAY N D. Method for registration of 3-D shapes [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 1992, 14(2):239-256. |
32 | LOWE D G. Object recognition from local scale-invariant features[C]∥Proceedings of the Seventh IEEE International Conference on Computer Vision. Piscataway: IEEE Press, 2002: 1150-1157. |
33 | SUMIKURA S, SHIBUYA M, SAKURADA K. OpenVSLAM: A versatile visual SLAM framework[C]∥Proceedings of the 27th ACM International Conference on Multimedia. New York: ACM, 2019: 2292-2295. |
34 | CAMPOS C, ELVIRA R, RODRíGUEZ J J G, et al. ORB-SLAM3: An accurate open-source library for visual, visual?inertial, and multimap SLAM[J]. IEEE Transactions on Robotics, 2021, 37(6): 1874-1890. |
35 | YOUSIF K, TAGUCHI Y, RAMALINGAM S. MonoRGBD-SLAM: Simultaneous localization and mapping using both monocular and RGBD cameras[C]∥ 2017 IEEE International Conference on Robotics and Automation (ICRA). Piscataway: IEEE Press, 2017: 4495-4502. |
36 | ZHANG Y T, SONG J M, DING Y, et al. FSD-BRIEF: A distorted BRIEF descriptor for fisheye image based on spherical perspective model[J]. Sensors (Basel, Switzerland), 2021, 21(5): 1839. |
37 | ROSTEN E, DRUMMOND T. Machine learning for high-speed corner detection[M]∥Computer Vision?ECCV 2006. Berlin: Springer, 2006: 430-443. |
38 | MIKOLAJCZYK K, SCHMID C. A performance evaluation of local descriptors[C]∥ 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Proceedings. Piscataway: IEEE Press, 2003: 1-7. |
39 | URBAN S, LEITLOFF J, HINZ S. MLPnP - A real-time maximum likelihood solution to the perspective-n-point problem[DB/OL].arXiv preprint: 1607.08112, 2016. |
/
〈 |
|
〉 |