特征稀疏场景下基于标签的车辆视觉SLAM

doi:10.19562/j.chinasae.qcgc.2023.09.004

Abstract

Abstract:

In the simultaneous localization and mapping of intelligent vehicles， the visual feature point method estimates the vehicle’s pose through extraction and matching of feature points. However， when the environment lacks texture or dynamic changes， due to sparse scene features and poor stability， localization by natural features possibly declines in accuracy or even fails. Adding visual tags in the environment can effectively solve the problem of feature sparsity. But the localization methods based on visual tags highly rely on manual calibration， and the poses often jitter due to perspective changes， which affects the precision of localization. Therefore， this paper proposes a tag-based vehicle visual SLAM method， which makes full use of tag information， introduces in internal and external corner constraints to reduce the pose jitter of the tag and establishes a low drift， globally consistent tag map with the visual odometer. The vehicle pose estimated by tags and the tag map are jointly optimized in localization to build a low-cost and highly robust visual SLAM system. The test results show that the proposed method with internal and external corner constraints effectively reduces the pose jitter of the tag， improves the mapping accuracy by more than 60% and the accuracy of localization by more than 30%， which significantly increases the accuracy and robustness of tag-based localization and is conducive to the safe operation of intelligent vehicles.

Key words: intelligent driving, simultaneous localization and mapping, image feature, visual tag, map consistency

Hongmao Qin,Guoli Shen,Yunshui Zhou,Shengjie Huang,Xiaohui Qin,Guotao Xie,Rongjun Ding. Tag-Based Vehicle Visual SLAM in Sparse Feature Scenes[J].Automotive Engineering, 2023, 45(9): 1543-1552.

Figures/Tables 21

组别	无内外角点约束方法/m				本文方法/m
组别	$L 0 ~ 5$	$L 6 ~ 15$	$L 16 ~ 21$	$L 22 ~ 31$	$L 0 ~ 5$	$L 6 ~ 15$	$L 16 ~ 21$	$L 22 ~ 31$
1	5.879 161	9.810 398	6.124 863	9.718 546	6.150 321	9.801 872	6.107 348	9.750 01
2	5.877 176	9.818 072	6.073 152	9.707 269	6.125 838	9.827 705	6.105 565	9.721 349
3	6.093 083	9.684 403	6.110 368	9.656 458	6.131 34	9.813 378	6.105 468	9.761 417
4	5.984 672	9.762 619	6.105 723	9.572 264	6.115 821	9.792 772	6.147 561	9.801 924
5	5.948 602	9.742 475	6.092 055	9.645 983	6.131 554	9.821 162	6.161 991	9.836 651
均值	5.956 539	9.763 593	6.101 232	9.661 016	6.130 975	9.811 378	6.125 587	9.774 27
Δ_max	0.136 544	0.079 19	0.028 08	0.088 751	0.019 346	0.018 606	0.021 974	0.052 921

References 24

1	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. IEEE， 2007： 225-234.
2	MUR-ARTAL R， MONTIEL J M M， TARDOS J D. ORB-SLAM： a versatile and accurate monocular SLAM system［J］. IEEE Transactions on Robotics， 2015， 31（5）： 1147-1163.
3	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.
4	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.
5	BANIK P P， SAHA R， KIM K D. Contrast enhancement of low-light image using histogram equalization and illumination adjustment［C］. 2018 International Conference on Electronics， Information， and Communication （ICEIC）. IEEE， 2018： 1-4.
6	WANG X， CHRISTIE M， MARCHAND E. Optimized contrast enhancements to improve robustness of visual tracking in a SLAM relocalisation context［C］. 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems （IROS）. IEEE， 2018： 103-108.
7	PUMAROLA A， VAKHITOV A， AGUDO A， et al. PL-SLAM： real-time monocular visual SLAM with points and lines［C］. 2017 IEEE International Conference on Robotics and Automation （ICRA）. IEEE， 2017： 4503-4508.
8	HE Y， ZHAO J， GUO Y， et al. Pl-vio： tightly-coupled monocular visual-inertial odometry using point and line features［J］. Sensors， 2018， 18（4）： 1159.
9	SUN Q， YUAN J， ZHANG X， et al. Plane-Edge-SLAM： seamless fusion of planes and edges for SLAM in indoor environments［J］. IEEE Transactions on Automation Science and Engineering， 2020， 18（4）： 2061-2075.
10	CHEN J， GAO Y， LI S. Real-time apriltag inertial fusion localization for large indoor navigation［C］. 2020 Chinese Automation Congress （CAC）. IEEE， 2020： 6912-6916.
11	CHAKRABORTY K， DEEGAN M， KULKARNI P， et al. JORB-SLAM： a jointly optimized multi-robot visual SLAM［J/OL］.http：//um-mobrob-t12-w19.github.io/docs/report.pdf.
12	PFROMMER B， DANIILIDIS K. TagSLAM： robust slam with fiducial markers［J］. arXiv preprint arXiv：， 2019.
13	ROOS-HOEFGEEST S， GARCIA I A， GONZALEZ R C. Mobile robot localization in industrial environments using a ring of cameras and ArUco markers［C］. IECON 2021–47th Annual Conference of the IEEE Industrial Electronics Society. IEEE， 2021： 1-6.
14	GUAN K， MA L， TAN X， et al. Vision-based indoor localization approach based on SURF and landmark［C］. 2016 International Wireless Communications and Mobile Computing Conference （IWCMC）. IEEE， 2016： 655-659.
15	OLSON E APRILTAG. A robust and flexible visual fiducial system［C］. 2011 IEEE International Conference on Robotics and Automation. IEEE， 2011： 3400-3407.
16	WANG J， OLSON E. AprilTag 2： efficient and robust fiducial detection［C］. 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems （IROS）. IEEE， 2016： 4193-4198.
17	KROGIUS M， HAGGENMILLER A， OLSON E. Flexible layouts for fiducial tags［C］. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems （IROS）. IEEE， 2019： 1898-1903.
18	SHI J. Good features to track［C］. 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 1994： 593-600.
19	QIN Xiaohui， WANG Zhewen， PANG Tao， et al. Vehicle positioning method based on tight coupling of vehicle model in enclosed environments［J］. Automotive Engineering， 2021， 43（9）： 1328-1335.
20	GUO Z， FU Ｑ， QUAN Q. Pose estimation for multicopters based on monocular vision and AprilTag［C］. 2018 37th Chinese Control Conference （CCC）. IEEE， 2018： 4717-4722.
21	NIU Guochen， FENG Ning， WANG Yu. Multi-layer map construction for outdoor unmanned vehicles based on graph optimization［J］. Automotive Engineering， 2022， 44（2）： 199-207.
22	HONG W， XIA H， AN X， et al. Natural landmarks based localization algorithm for indoor robot with binocular vision［C］. 2017 29th Chinese Control And Decision Conference （CCDC）. IEEE， 2017： 3313-3318.
23	WANG Zihao， CAI Yingfeng， WANG Hai， et al. Surrounding multi-target trajectory prediction method based on monocular visual motion estimation［J］. Automotive Engineering， 2022， 44（9）： 1318-1326.
24	STRISCIUGLIO N， VALLINA M L， PETKOV N， et al. Camera localization in outdoor garden environments using artificial landmarks［C］. 2018 IEEE International Work Conference on Bioinspired Intelligence （IWOBI）. IEEE， 2018： 1-6.

[1]	HAO Han, WANG Si-南, LI Xiao, LIU Zong-Wei, ZHAO Fu-Quan. [J]. , 2017, 39(1): 1 -8 .
[2]	CAO Li-Bo, HU Yuan, YAN Ling-Bo, PENG Yu, SHI Xiang-南. [J]. , 2017, 39(2): 174 -180 .
[3]	Chen-Xin, FENG Xiao, SHEN Chuan-Liang, WANG Shuo, HU Cui-Song, LI Yan-Yang, YANG Chang-Hai-　. [J]. , 2017, 39(2): 206 -213 .
[4]	LI Jun, SU Yan-Zhao, KUI Han-Bing, HU Ming-Hui, ZHANG Sheng-Gen, LIU Heng-Shuo. [J]. , 2017, 39(3): 296 -303 .
[5]	. Design and Implementation of Electric Vehicle Drive Motor Control Based on AUTOSAR[J]. , 0, (): 0 .
[6]	LIU Wei-Da, WANG Tie, SHEN Jin-Xian, YI Chen-Yang. [J]. , 2017, 39(3): 323 -327 .
[7]	. Research on Defrosting Prediction Method based on Evaporating Temperature in Mobile Air-Conditioning System[J]. , 0, (): 0 .
[8]	. Simulation and Test of Vibration and Noise for the Drive Axle[J]. , 0, (): 0 .
[9]	ZHOU Min, ZHANG Jie, ZHENG Min-Yi, ZHANG Nong, ZHANG Bang-Ji. [J]. , 2017, 39(4): 447 -456 .
[10]	CHU Liang, LIU Da-Liang, LIU Hong-Wei, CAI Jian-Wei, ZHAO Di. [J]. , 2017, 39(4): 471 -479 .

组别	1	2	3	4	5
ORB-SLAM3	35.322	34.455	33.664	33.581	34.216
本文方法	46.537	46.358	47.876	46.107	45.764
Δ	11.215	11.903	14.212	12.526	11.548

组别		1	2	3	4	5
基于里程计	ORB-SLAM3/m	0.787	0.436	0.476	0.673	0.516
	本文方法/m	0.497	0.293	0.189	0.392	0.264
	提升率/%	36.85	32.79	60.29	41.75	48.83
基于地图	ORB-SLAM3/m	0.413	0.430	0.468	0.692	0.507
	本文方法/m	0.291	0.236	0.254	0.495	0.294
	提升率/%	29.54	45.12	45.72	28.47	42.21

模块	特征提取	跟踪	建图	总计
ORB-SLAM3	16.398	5.502	16.069	37.969
本文方法	41.631	5.519	15.907	63.057

Tag-Based Vehicle Visual SLAM in Sparse Feature Scenes

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 21

References 24

Related Articles 8

Metrics

Comments

Recommended 10

[1]	Jiangkun Li,Weiwen Deng,Bingtao Ren,Wenqi Wang,Juan Ding. Automatic Driving Edge Scene Generation Method Based on Scene Dynamics and Reinforcement Learning [J]. Automotive Engineering, 2022, 44(7): 976-986.
[2]	Bin Tang,Zhanxiang Xu,Haobin Jiang,Yingfeng Cai,Zitian Hu,Zhengyi Yang. Trajectory Planning of Intelligent Vehicles in Lane Change for Collision Avoidance Based on Segmented Optimization [J]. Automotive Engineering, 2022, 44(6): 831-841.
[3]	Shengjie Huang,Manjiang Hu,Yunshui Zhou,Zhouping Yin,Xiaohui Qin,Yougang Bian,Qianqian Jia. Vehicle Visual SLAM in Dynamic Scenes Based on Semantic Segmentation and Motion Consistency Constraints [J]. Automotive Engineering, 2022, 44(10): 1503-1510.
[4]	Xiaolong Zhang,Kun Jiang,Kai Sun,Maolin Liu,Zhenlei Shao,Pengchuan Xiao,Diange Yang. Intrinsic Parameters Modeling for LiDAR and Point Cloud Correction Method [J]. Automotive Engineering, 2021, 43(8): 1228-1237.
[5]	Jinghua Guo,Wenchang Li,Yugong Luo,Tao Chen,Keqiang Li. Driver Car⁃Following Model Based on Deep Reinforcement Learning [J]. Automotive Engineering, 2021, 43(4): 571-579.
[6]	Liu Zhiqiang, Han Jingwen, Ni Jie. Study on Multi-vehicle Coordinated Lane Change Strategy Under Network Conditions [J]. Automotive Engineering, 2020, 42(3): 299-306.
[7]	Li Yayong, Cai Yingfeng, Chen Long, Sun Xiaoqiang, He Youguo, Zhang Yunshun. ACC Method Considering Driving State of Front and Rear Vehicles [J]. Automotive Engineering, 2019, 41(8): 865-871.
[8]	Zhang Jinhui, Li Keqiang, Luo Yugong, Zhang Shuwei & Li Hong. Prediction of Preceding Car Motion Under Car-following Scenario in the Internet of Vehicle Based on Bayesian Network [J]. , 2019, 41(3): 245-251.