VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator
Open accessThe DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Abstract
One camera and one low-cost inertial measurement unit (IMU) form a monocular visual-inertial system (VINS), which is the minimum sensor suite (in size, weight, and power) for the metric six degrees-of-freedom (DOF) state estimation. In this paper, we present VINS-Mono: a robust and versatile monocular visual-inertial state estimator. Our approach starts with a robust procedure for estimator initialization. A tightly coupled, nonlinear optimization-based method is used to obtain highly accurate visual-inertial odometry by fusing preintegrated IMU measurements and feature observations. A loop detection module, in combination with our tightly coupled formulation, enables relocalization with minimum computation. We additionally perform 4-DOF pose graph optimization to enforce the global consistency. Furthermore, the proposed system can reuse a map by saving and loading it in an efficient way. The current and previous maps can be merged together by the global pose graph optimization. We validate the performance of our system on public datasets and real-world experiments and compare against other state-of-the-art algorithms. We also perform an onboard closed-loop autonomous flight on the microaerial-vehicle platform and port the algorithm to an iOS-based demonstration. We highlight that the proposed work is a reliable, complete, and versatile system that is applicable for different applications that require high accuracy in localization. We open source our implementations for both PCs (https://github.com/HKUST-Aerial-Robotics/VINS-Mono) and iOS mobile devices (https://github.com/HKUST-Aerial-Robotics/VINS-Mobile).
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In terms of vision-IMU-GPS fusion, VINS-Fusion has become a mainstream solution for large-scale, high-precision localization and is widely used in UAVs and mobile robots.”
From Multimodal fusion and vision–language models: A survey for robot vision (2026) · cites this paper as [151]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Multiple View Geometry in Computer Vision
Hartley, R., & Zisserman, A. (2004) · doi:10.1017/cbo9780511811685 · Cited by 21,070
Are we ready for autonomous driving? The KITTI vision benchmark suite
Geiger, A., Lenz, P., & Urtasun, R. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248074 · Cited by 14,897 · cited 3×
“Relative pose error [42] in MH_05_difficult.”
MethodInfluential(1981) · Cited by 11,963
Shi, J., & Tomasi (1994) · Proceedings of IEEE Conference on Computer Vision and Pattern Recognition CVPR-94 · doi:10.1109/cvpr.1994.323794 · Cited by 7,724 · cited 1×
“features are detected [32] to maintain a minimum number (100– 300) of features in each image.”
BackgroundRobust estimation of a location parameter
Huber, P. J. (1964) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177703732 · Cited by 7,228
ORB-SLAM: A Versatile and Accurate Monocular SLAM System
Mur-Artal, R., Montiel, J. M. M., & Tardos, J. D. (2015) · IEEE Transactions on Robotics · doi:10.1109/tro.2015.2463671 · Cited by 6,778 · cited 7×
“ORB-SLAM [4] is able to close loops and reuse the map, which takes advantage of Bag-of-World [6].”
BackgroundMethodInfluentialParallel Tracking and Mapping for Small AR Workspaces
Klein, G., & Murray, D. (2007) · 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2007.4538852 · Cited by 4,346
Bundle Adjustment — A Modern Synthesis
Triggs, B., McLauchlan, P. F., Hartley, R. I., & Fitzgibbon, A. W. (2000) · Lecture Notes in Computer Science · doi:10.1007/3-540-44480-7_21 · Cited by 3,988 · cited 1×
“Finally, a global full bundle adjustment [36] is applied to minimize the total reprojection error of all feature observations.”
MethodA benchmark for the evaluation of RGB-D SLAM systems
Sturm, J., Engelhard, N., Endres, F., Burgard, W., & Cremers, D. (2012) · 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2012.6385773 · Cited by 3,976 · cited 3×
“TABLE I RMSE [43] IN EUROC DATASETS IN METERS”
BackgroundMethodInfluentialLSD-SLAM: Large-scale direct monocular SLAM
Engel, J., Schöps, T., & Cremers, D. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10605-2_54 · Cited by 3,851 · cited 3×
“Noticeable approaches include PTAM [1], SVO [2], LSD-SLAM [3], DSO [5], and ORB-SLAM [4].”
BackgroundMethod
Cited by
Items 201–300 of 1,001
A Survey of Visual SLAM in Dynamic Environment: The Evolution From Geometric to Semantic Approaches
Wang, Y., Tian, Y., Chen, J., Xu, K., & Ding, X. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3420374 · Cited by 144
Observability Analysis of Aided INS With Heterogeneous Features of Points, Lines, and Planes
Yang, Y., & Huang, G. P. (2019) · IEEE Transactions on Robotics · doi:10.1109/tro.2019.2927835 · Cited by 97
“, [4], [5], [14]–[17]) are developed based on point features, the observability analysis is performed primarily using point measurements.”
Krul, S., Pantos, C., Frangulea, M., & Valente, J. R. P. (2021) · Drones · doi:10.3390/drones5020041 · Cited by 91
Exploring Event Camera-Based Odometry for Planetary Robots
Mahlknecht, F., Gehrig, D., Nash, J., Rockenbauer, F. M., Morrell, B. J., Delaune, J. H., & Scaramuzza, D. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3187826 · Cited by 102
Underwater Robots and Key Technologies for Operation Control
Sun, L., Wang, Y., Hui, X., Ma, X., Bai, X., & Tan, M. H. (2024) · Cyborg and Bionic Systems · doi:10.34133/cbsystems.0089 · Cited by 92
“In 2018, Burguera and Bonin-Font [85] proposed a multi-session monocular approach that adopts the local single-session, global multi-session loop detectors and the optimizer based on the Iterated Extension Kalman Filter (I-EKF).”
Degenerate Motion Analysis for Aided INS With Online Spatial and Temporal Sensor Calibration
Yang, Y., Geneva, P., Eckenhoff, K., & Huang, G. P. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2893803 · Cited by 95
End-to-End Learning Framework for IMU-Based 6-DOF Odometry
Lima, J. P., Uchiyama, H., & Taniguchi, R.-I. (2019) · Sensors · doi:10.3390/s19173777 · Cited by 78
“There are several 6-DOF visual-inertial odometry (VIO) methods available, such as VINS-MONO [9] and OKVIS [10], and their performance was summarized in [1].”
MethodOleynikova, H., Lanegger, C., Taylor, Z., Pantic, M., Millane, A., Siegwart, R., & Nieto, J. I. (2020) · Journal of Field Robotics · doi:10.1002/rob.21950 · Cited by 87
“e, even on the on-board CPU, and using direct photometric error makes the method resistant to motion blur. In our experience, Rovio is comparably accurate to other methods such as Okvis and VINS-mono [14], but more robust under real conditions. However, Rovio only gives us the odometry in the sensor frame, which for o…”
MethodFast-Racing: An Open-Source Strong Baseline for $\mathrm{SE}(3)$ Planning in Autonomous Drone Racing
Han, Z., Wang, Z., Pan, N., Lin, Y., Xu, C., & Gao, F. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3113976 · Cited by 76
From SLAM to Situational Awareness: Challenges and Survey
Bavle, H., Sánchez-López, J. L., Cimarelli, C., Tourani, A., & Voos, H. (2023) · Sensors · doi:10.3390/s23104849 · Cited by 108
Zhang, Z., Zhao, J., Huang, C., & Li, L. (2022) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2022.3173662 · Cited by 83
“An alternative approach for sensor fusion is the recently developed factor graph optimization (FGO) [14], [22], [23].”
MethodAerial orthoimage generation for UAV remote sensing: Review
Zhang, J., Xu, S., Zhao, Y., Sun, J., Xu, S., & Zhang, X. (2022) · Information Fusion · doi:10.1016/j.inffus.2022.08.007 · Cited by 88
Jiang, G., Yin, L., Jin, S., Tian, C., Ma, X., & Ou, Y. (2019) · Applied Sciences · doi:10.3390/app9102105 · Cited by 85
TUM-VIE: The TUM Stereo Visual-Inertial Event Dataset
Klenk, S., Chui, J., Demmel, N., & Cremers, D. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636728 · Cited by 91
“We provide evaluation for Basalt [17] and VINS-Fusion [18, 19, 20].”
MethodDynaNet: Neural Kalman Dynamical Model for Motion Estimation and Prediction
Chen, C., Lu, C. X., Wang, B., Trigoni, N., & Markham, A. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3112460 · Cited by 85
Real-Time Visual–Inertial SLAM Based on Adaptive Keyframe Selection for Mobile AR Applications
Piao, J.-C., & Kim, S. (2019) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2913324 · Cited by 83
“The mean translation RMSE values of the AVIO-based SLAM, OKVIS_mono, OKVIS_stereo, and VINS are 0.059 m, 0.225 m, 0.122 m, and 0.114 m, respectively.”
MethodResultInfluentialCNN-Based Dense Monocular Visual SLAM for Real-Time UAV Exploration in Emergency Conditions
Steenbeek, A., & Nex, F. (2022) · Drones · doi:10.3390/drones6030079 · Cited by 80
“The most common solution is Visual Inertial SLAM (VI-SLAM), where IMU measurements serve as odometry and are fused with the SLAM measurements, often utilizing a Kalman filter [29].”
MethodA Novel Lidar-Assisted Monocular Visual SLAM Framework for Mobile Robots in Outdoor Environments
Yin, J., Luo, D., Yan, F., & Zhuang, Y. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3190031 · Cited by 85
ASRO-DIO: Active Subspace Random Optimization Based Depth Inertial Odometry
Zhang, J., Tang, Y., Wang, H., & Xu, K. (2022) · IEEE Transactions on Robotics · doi:10.1109/tro.2022.3208503 · Cited by 88
“where P (·) and Q(·) are the mid-point propagation function introduced in [15], and Eak and E B gk are included in our state xk and represent the average IMU measurement error from tk−1 to tk.”
BackgroundMethodInfluentialEPLF-VINS: Real-Time Monocular Visual-Inertial SLAM With Efficient Point-Line Flow Features
Xu, L., Yin, H., Shi, T., JIANG, D. M., & Huang, B. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3231983 · Cited by 84
“Visual-inertial SLAM navigation systems (VINS), which are low-cost and lightweight compared with other multisensor fusion solutions, have received much attention from researchers [1], [2], [3].”
BackgroundZhao, Y., Yan, L., Xie, H., Dai, J., & Wei, P. (2023) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2023.3285921 · Cited by 74
“FOV sensor and use [28] to provide the quadrotor state.”
Integrated Polarized Skylight Sensor and MIMU With a Metric Map for Urban Ground Navigation
Fan, C., Hu, X., He, X., Zhang, L., & Lian, J. (2017) · IEEE Sensors Journal · doi:10.1109/jsen.2017.2786404 · Cited by 74
“The performance of proposed method is compared with three methods: the monocular visual odometry (VO) [4], the visual-inertial navigation system (VINS) [6] and the integrated method of the PS and MIMU.”
BackgroundMethodInfluentialA Novel Dual-Lidar Calibration Algorithm Using Planar Surfaces
Jiao, J., Liao, Q., Zhu, Y., Liu, T., Yu, Y., Fan, R., Wang, L., & Liu, M. (2019) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2019.8814136 · Cited by 81
“Generally, motion-based methods can work for a variety of configurations and can be integrated into several SLAM systems [23].”
BackgroundGood Feature Matching: Toward Accurate, Robust VO/VSLAM With Low Latency
Zhao, Y., & Vela, P. A. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.2964138 · Cited by 87
“The RMSE of the VINS methods, labeled SVOMSF[69] and VIMono[70], are obtained by”
MethodLearned Visual Navigation for Under-Canopy Agricultural Robots
Sivakumar, A. N. V., Modi, S., Gasparino, M. V., Ellis, C., Velasquez, A. E. B., Chowdhary, G. V., & Gupta, S. (2021) · Robotics: Science and Systems · doi:10.15607/rss.2021.xvii.019 · Cited by 103
“To compare with a feature matching based VIO algorithm, Vins-fusion was used as the baseline [50].”
BackgroundMethodRobust Place Recognition using an Imaging Lidar
Shan, T., Englot, B., Duarte, F., Ratti, C., & Rus, D. L. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9562105 · Cited by 90
“For example, [4] introduces a twostep geometric validation method, which triangulates visual features, to verify the candidates.”
MethodClimbing robots for manufacturing
Tao, B., Gong, Z., & Ding, H. (2023) · National Science Review · doi:10.1093/nsr/nwad042 · Cited by 79
Optimization-Based Visual-Inertial SLAM Tightly Coupled with Raw GNSS Measurements
Liu, J., Gao, W., & Hu, Z. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9562013 · Cited by 72
“Our approach is evaluated on 4 sequences collected in San Francisco, because the other 3 sequences in San Francisco contain tunnels, which are very tough scenerios for both camera and GNSS signals, and VINS-mono [16] evaluation results on only the above 4 sequences are reported in the paper [23].”
BackgroundMethodResultInfluentialVisual-Inertial Odometry with Point and Line Features
Yang, Y., Geneva, P., Eckenhoff, K., & Huang, G. P. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8967905 · Cited by 79
“By contrast, indirect systems typically extract geometric features from the pixels and track their motion across the image plane with the most common being points [5–7] and/or lines [8–13].”
BackgroundAutonomous Cave Surveying With an Aerial Robot
Tabib, W., Goel, K., Yao, J., Boirum, C., & Michael, N. (2021) · IEEE Transactions on Robotics · doi:10.1109/tro.2021.3104459 · Cited by 71
“C. Hardware Experiments 1) Visual-Inertial Navigation and Control: State estimates are computed from IMU and downward-facing camera observations via VINS-Mono [50], a tightly-coupled visual-inertial odometry framework that jointly optimizes vehicle motion, feature locations, and IMU biases over a sliding window of mon…”
InfluentialMarshall, J. A., Sun, W., & L’Afflitto, A. (2021) · Annual Reviews in Control · doi:10.1016/j.arcontrol.2021.10.013 · Cited by 85
“Sensor fusion, such as the combination of visual and IMU information (Qin, Li, & Shen, 2018b; Wang, Yang, Cao, Xu, & Lin, 2018), the fusion of camera and LiDAR sensor measurement (Shin, Park, & Kim, 2020), or the inclusion of semantic information (Atanasov, Bowman, Daniilidis, & Pappas, 2018), also…”
PL-EVIO: Robust Monocular Event-Based Visual Inertial Odometry With Point and Line Features
Guan, W., Chen, P., Xie, Y., & Lu, P. (2023) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2023.3324365 · Cited by 98
“raw image from DAVIS346 to run the VINS-MONO [36], PL-VINS [4], and ORB-SLAM3 [42], as image-based comparisons.”
Lin, J., & Zhang, F. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3456473 · Cited by 101
“The scale issue in pure visual SLAM could also be addressed by fusing inertial sensor data, as demonstrated by VINS-Mono [30], Kimera [31], and ORB-SLAM3 [32], which achieve high-accuracy localization by fusing IMU measurements and image features in a sliding window bundle adjustment optimization.”
MethodLiDAR, IMU, and camera fusion for simultaneous localization and mapping: a systematic review
Fan, Z., Zhang, L., Wang, X., Shen, Y., & Deng, F. Y. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11187-w · Cited by 110
A Lightweight and Accurate Localization Algorithm Using Multiple Inertial Measurement Units
Zhang, M., Xu, X., Chen, Y., & Li, M. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.2969146 · Cited by 73
Monocular Camera Localization in Prior LiDAR Maps with 2D-3D Line Correspondences
Yu, H., Zhen, W., Yang, W. T., Zhang, J., & Scherer, S. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341690 · Cited by 74
“Dataset VINS-Mono VINS-Mono 2D-3D (odom)[2] (loop)[2] matching V1 01 easy 0.”
BackgroundMethodResultInfluentialKhatiri, S., Panichella, S., & Tonella, P. (2023) · IEEE International Conference on Software Testing, Verification and Validation (ICST) · doi:10.1109/icst57152.2023.00034 · Cited by 91
“State estimation recreates the drone state in the environment and enables navigation and autonomous movement [39], while mapping strategies compute obstacle distances, to create a model of the surrounding area [40].”
BackgroundAn Overview of Key SLAM Technologies for Underwater Scenes
Wang, X., Fan, X., Shi, P., Ni, J., & Zhou, Z. (2023) · Remote Sensing · doi:10.3390/rs15102496 · Cited by 80
“Generally, the main visual–inertial localization methods can be divided into two categories: a class of Kalman filter based methods, such as MSCKF [65] and ROVIO [66]; and a class of graph-based optimization methods, for instance, OKVIS [67] and VINS-Mono [68].”
MethodIntroducing autonomous aerial robots in industrial manufacturing
Pérez-Grau, F. J., Dios, J. R. M., Paneque, J. L., Acevedo, J. J., Torres‐González, A., Viguria, A., Astorga, J. R., & Ollero, A. (2021) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2021.06.008 · Cited by 69
Tightly-Coupled Monocular Visual-Odometric SLAM Using Wheels and a MEMS Gyroscope
Quan, M., Piao, S., Tan, M., & Huang, S.-S. (2019) · IEEE Access · doi:10.1109/access.2019.2930201 · Cited by 74
“We also performed this sequence on the state-of-the-art and open-source monocular VSLAM system ORB-SLAM [14] and monocular VISLAM system VINS-MONO [6], [26] for comparison.”
BackgroundMethodLightweight 3-D Localization and Mapping for Solid-State LiDAR
Wang, H., Wang, C., & Xie, L. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3060392 · Cited by 71
“SLAM system such as visual SLAM [6], [7] and visual inertial SLAM [8], [9].”
An RGB-D Camera Based Visual Positioning System for Assistive Navigation by a Robotic Navigation Aid
Zhang, H., Jin, L., & Ye, C. (2021) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2021.1004084 · Cited by 75
“Related Work in VIO Existing VIO methods can be classified into two categories, namely loosely-coupled [19]–[21] and tightly-coupled [22], [16]–[18].”
BackgroundMethodInfluentialA LiDAR/Visual SLAM Backend with Loop Closure Detection and Graph Optimization
Chen, S., Zhou, B., Jiang, C., Xue, W., & Li, Q. (2021) · Remote Sensing · doi:10.3390/rs13142720 · Cited by 75
“Traditionally, SLAM technology is divided according to the employed sensors, i.e., visual SLAM employs cameras as the sensor, while LiDAR SLAM utilizes LiDAR as the sensor to scan the environment [6–12].”
MethodAn Adaptive Weighting Strategy for Multisensor Integrated Navigation in Urban Areas
Sun, R., Dai, Y., & Cheng, Q. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3256008 · Cited by 70
“VINS is a very advanced monocular VIO tightly coupled algorithm based on nonlinear optimization and sliding windows [12].”
MethodMVS‐SLAM: Enhanced multiview geometry for improved semantic RGBD SLAM in dynamic environment
Islam, Q. U., Ibrahim, H., Chin, P. K., Lim, K., & Abdullah, M. Z. (2023) · Journal of Field Robotics · doi:10.1002/rob.22248 · Cited by 74
Li, J., Yuan, S., Cao, M., Nguyen, T., Cao, K., & Xie, L. (2024) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2024.04.004 · Cited by 74
MCD: Diverse Large-Scale Multi-Campus Dataset for Robot Perception
Nguyen, T., Yuan, S., Nguyen, T. H., Yin, P., Cao, H., Xie, L., Wozniak, M. K., Jensfelt, P., Thiel, M., Ziegenbein, J., et al. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.02105 · Cited by 110
“Three SOTA visual-inertial SLAM methods are tested against MCD: VINS [33], KIMERA [35], and OPENVINS [15].”
SubT-MRS Dataset: Pushing SLAM Towards All-weather Environments
Zhao, S., Gao, Y., Wu, T., Singh, D., Jiang, R., Sun, H., Sarawata, M., Qiu, Y., Whittaker, W., Higgins, I., et al. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.02137 · Cited by 84
Wang, S., & Ahmad, N. S. (2024) · IEEE Access · doi:10.1109/access.2024.3519874 · Cited by 91
“Qin et al. [156]’s Visual-Inertial Navigation System (VINS-Mono) supports online relocalization and four-degree-of-freedom global pose graph optimization, although tracking performance might degrade under poor image quality or extreme lighting conditions.”
DRE-SLAM: Dynamic RGB-D Encoder SLAM for a Differential-Drive Robot
Yang, D., Bi, S., Wang, W., Chang, Y., Wang, W., Qi, X., & Cai, Y. (2019) · Remote Sensing · doi:10.3390/rs11040380 · Cited by 75
“The VI-SLAM methods, such as VINS-Mono [5], are robust in dynamic environments.”
MethodTrifo-VIO: Robust and Efficient Stereo Visual Inertial Odometry Using Points and Lines
Zheng, F., Tsai, G., Zhang, Z., Liu, S., Chu, C.-C., & Hu, H. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8594354 · Cited by 75
Liu, J., Gao, W., & Hu, Z. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8967607 · Cited by 77
“Similar to [2] and [3], our nonlinear optimization is performed in sliding window with partial marginalization.”
BackgroundMethodInfluentialTopology Aware Object-Level Semantic Mapping Towards More Robust Loop Closure
Lin, S., Wang, J., Xu, M., Zhao, H., & Chen, Z. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3097242 · Cited by 73
“Traditional appearance-based loop detection methods [2], [3] model the environment as a set of images captured in the process of mapping, and scene is recognized according to the similarity between images, which is performed by local feature extraction and matching [4].”
MethodA comprehensive overview of core modules in visual SLAM framework
Cai, D., Li, R., Hu, Z., Lu, J., Li, S., & Zhao, Y. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127760 · Cited by 75
Learning vision-based agile flight via differentiable physics
Zhang, Y., Hu, Y., Song, Y., Zou, D., & Lin, W. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01048-0 · Cited by 146
“The objective of our optimization problem is to find the optimal policy parameters θ * by minimizing the total loss using gradient descent: where γ is the learning rate.”
Relocalization, Global Optimization and Map Merging for Monocular Visual-Inertial SLAM
Qin, T., Li, P., & Shen, S. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8460780 · Cited by 75
Self-citationUAV/UGV Autonomous Cooperation: UAV assists UGV to climb a cliff by attaching a tether
Miki, T., Khrapchenkov, P., & Hori, K. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794265 · Cited by 77
“1) VIO: As also discussed by [22], there are several opensource libraries available for VIO [23, 24, 25, 26].”
BackgroundInfluentialVisual SLAM in dynamic environments based on object detection
Ai, Y., Rui, T., Yang, X., He, J., Fu, L., Li, J., & Lu, M. (2020) · Defence Technology · doi:10.1016/j.dt.2020.09.012 · Cited by 74
Bridge Inspection Using Unmanned Aerial Vehicle Based on HG-SLAM: Hierarchical Graph-Based SLAM
Jung, S., Choi, D., Song, S., & Myung, H. (2020) · Remote Sensing · doi:10.3390/rs12183022 · Cited by 68
Qiu, K., Qin, T., Pan, J., Liu, S., & Shen, S. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.3033698 · Cited by 77
“M ULTISENSOR fusion is fundamental to various robotic applications based on accurate environment perception, such as simultaneous localization and mapping (SLAM) [1] and dynamic environment perception [2], [3].”
BackgroundSelf-citation3D ego-Motion Estimation Using low-Cost mmWave Radars via Radar Velocity Factor for Pose-Graph SLAM
Park, Y. S., Shin, Y.-S., Kim, J.-W., & Kim, A. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3099365 · Cited by 82
“Recently, significant advances have been made in visual odometry, which estimates 3D trajectory and motion using sequential images from cameras and LiDARs [1, 2, 3, 4].”
BackgroundSDV-LOAM: Semi-Direct Visual–LiDAR Odometry and Mapping
Yuan, Z., Wang, Q., Cheng, K., Hao, T., & Yang, X. Y. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3262817 · Cited by 88
“In general, vision-based odometry/SLAM systems can be categorized into three classes: 1) feature-based methods [1], [25], [26], [27], [28] which rely on local image features for pose tracking, 2) direct methods [2], [3], [6], [7], [29] which directly use raw sensor measurements (e.g., intensity…”
MethodSensor Fusion of GNSS and IMU Data for Robust Localization via Smoothed Error State Kalman Filter
Yin, Y., Zhang, J., Guo, M., Ning, X., Wang, Y., & Lu, J. (2023) · Sensors · doi:10.3390/s23073676 · Cited by 76
Efficient and Consistent Bundle Adjustment on Lidar Point Clouds
Liu, Z., Liu, X., & Zhang, F. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2023.3311671 · Cited by 88
“Compared to visual BA, which has been well-established in photogrammetry and played a fundamental role in various vital applications, including visual odometry (VO) [23], [24], [25], visual-inertial odometry (VIO) [26], [27], 3-”
MethodTracking 3-D Motion of Dynamic Objects Using Monocular Visual-Inertial Sensing
Qiu, K., Qin, T., Gao, W., & Shen, S. (2019) · IEEE Transactions on Robotics · doi:10.1109/tro.2019.2909085 · Cited by 72
“Both VINS and region-based BA are formulated as sliding window-based nonlinear optimization problems, and a unified representation of the sliding window structure is shown in Fig.”
BackgroundMethodInfluentialSelf-citationA Survey on Visual Navigation and Positioning for Autonomous UUVs
Qin, J., Li, M., Li, D., Zhong, J., & Yang, K. (2022) · Remote Sensing · doi:10.3390/rs14153794 · Cited by 65
“However, limited by various complex environments that may actually exist, such as dynamic targets, missing textures, complex light, possible inaccurate camera calibration, and inaccurate system modeling, it may lead to low positioning accuracy or even positioning failure.”
BackgroundA GNSS/INS/LiDAR Integration Scheme for UAV-Based Navigation in GNSS-Challenging Environments
Elamin, A., Abdelaziz, N., & El‐Rabbany, A. (2022) · Sensors · doi:10.3390/s22249908 · Cited by 71
EgoLocate: Real-time Motion Capture, Localization, and Mapping with Sparse Body-mounted Sensors
Yi, X., Zhou, Y., Habermann, M., Golyanik, V., Pan, S., Theobalt, C., & Xu, F. (2023) · ACM Transactions on Graphics · doi:10.1145/3592099 · Cited by 90
“Visual-inertial SLAM incorporates an IMU to the camera, which becomes more robust to fast camera motions and poor visual features, and can also determine the scale factor [Campos et al. 2021; Leutenegger et al. 2015; Mourikis et al. 2007; Mur-Artal and Tardós 2017b; Qin et al. 2018; Von Stumberg et al. 2018].”
BackgroundMethodInfluentialLiu, X., Wen, W., & Hsu, L. (2023) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2023.3323648 · Cited by 82
“The implementation of IMU pre-integration follows [38], the relevant theories are omitted here and interested readers can refer to [14], [38] for more information in detail.”
BackgroundSwitch-SLAM: Switching-Based LiDAR-Inertial-Visual SLAM for Degenerate Environments
Lee, J., Komatsu, R., Shinozaki, M., Kitajima, T., Asama, H., An, Q., & Yamashita, A. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3421792 · Cited by 79
“3) Visual Odometry: The method in [6] is adapted for our visual odometry submodule.”
BackgroundMethodInfluentialUnscented Kalman Filter on Lie Groups for Visual Inertial Odometry
Brossard, M., Bonnabel, S., & Barrau, A. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593627 · Cited by 80
“In this section, we compare the performances of the proposed S-UKF-LG and its full (Invariant-)EKF variant, that we call S-IEKF, with state-of-the-art VIO algorithms including S-MSCKF [2], OKVIS (stereo-optimization) [12] and VINS-MONO (monocular-optimization) [13], i.”
MethodVisual-SLAM Classical Framework and Key Techniques: A Review
Jia, G., Li, X., Zhang, D., Xu, W., Lv, H., Shi, Y. C., & Cai, M. (2022) · Sensors · doi:10.3390/s22124582 · Cited by 69
AirVO: An Illumination-Robust Point-Line Visual Odometry
Xu, K., Hao, Y., Yuan, S., Wang, C., & Xie, L. (2023) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros55552.2023.10341914 · Cited by 76
“Many of these features, e.g., ORB [13], FAST [14], and BRISK [15] are applied to VO and SLAM systems, e.g., ORB-SLAM [16], VINS-Mono [3], because of their balanced effectiveness and efficiency.”
BackgroundMethodInfluentialEdge-Assisted Multi-Robot Visual-Inertial SLAM With Efficient Communication
Liu, X., Wen, S., Zhao, J., Qiu, T. Z., & Zhang, H. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2024.3376427 · Cited by 76
“The VINS-Mono proposed by Qin et al. [33] is a powerful and universal monocular VI state estimator.”
A Survey on Autonomous and Intelligent Swarms of Uncrewed Aerial Vehicles (UAVs)
Du, Z., Luo, C., Min, G., Wu, J. J., Luo, C. L., Pu, J., & Li, S. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3569500 · Cited by 87
“IMU is an essential sensor for providing feedback con-trol in UAVs’ attitude loops, making visual-inertial odometry (VIO) highly suitable for UAV positioning [207], [208], [209].”
Efficient Trajectory Planning for High Speed Flight in Unknown Environments
Ryll, M., Ware, J., Carter, J., & Roy, N. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8793930 · Cited by 64
“It has been noted that this allows visual-inertial odometry systems to increase the number of trackable features between camera frames and reduce the discontinuities in position and rotation between state estimator outputs [9].”
BackgroundZhang, M., Zuo, X., Chen, Y., Liu, Y., & Li, M. (2021) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.3043970 · Cited by 75
“Since they are standard steps for sliding-window pose estimation [14], [26], we omit the details here.”
BackgroundMethodInfluentialSOFT2: Stereo Visual Odometry for Road Vehicles Based on a Point-to-Epipolar-Line Metric
Cvišić, I., Marković, I., & Petrovic, I. K. (2022) · IEEE Transactions on Robotics · doi:10.1109/tro.2022.3188121 · Cited by 82
“For reference, we also process every sequence with the publicly available ORB-SLAM2 [10], OV2SLAM [33], and VINS-FUSION [11], [43] implementations, and compare the results to SOFT2.”
MethodResultRT-TEE: Real-time System Availability for Cyber-physical Systems using ARM TrustZone
Wang, J., Li, A., Li, H., Lu, C., & Zhang, N. (2022) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp46214.2022.9833604 · Cited by 74
“To visualize the potential impact, we launched the attack against the control program of the drone, which is using VINS-Fusion [38] for localization.”
MethodSemantic localization on BIM-generated maps using a 3D LiDAR sensor
Yin, H., Lin, Z., & Yeoh, J. K. W. (2022) · Automation in Construction · doi:10.1016/j.autcon.2022.104641 · Cited by 82
“In robotics, a general localization approach is the Simultaneous Localization and Mapping (SLAM) system [5, 6], which achieves mapping and localization simultaneously using onboard sensors.”
MethodVoxfield: Non-Projective Signed Distance Fields for Online Planning and 3D Reconstruction
Pan, Y., Kompis, Y., Bartolomei, L., Mascaro, R., Stachniss, C., & Chli, M. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981318 · Cited by 85
“State estimation [22], TSDF and ESDF mapping with Voxfield and path planning [23] all run onboard the MAV.”
MethodBotanicGarden: A High-Quality Dataset for Robot Navigation in Unstructured Natural Environments
Yuan-zhi, L., Fu, Y., Qin, M., Xu, Y., Xu, B., Chen, F., Goossens, B., Sun, P. Z. H., Yu, H., Liu, C., et al. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3359548 · Cited by 91
“Famous SO/SLAM frameworks include VINS-Mono [13], ORB-SLAM [14], [15], LOAM and its extensions [16], [17], etc .”
MethodVIMO: Simultaneous Visual Inertial Model-Based Odometry and Force Estimation
Nisar, B., Foehn, P., Falanga, D., & Scaramuzza, D. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2918689 · Cited by 76
Du, H., Wang, W., Xu, C., Xiao, R., & Sun, C. (2020) · Sensors · doi:10.3390/s20030919 · Cited by 69
RD-VIO: Robust visual-inertial odometry for mobile augmented reality in dynamic environments
Li, J., Pan, X., Huang, G., Zhang, Z., Wang, N., Bao, H., & Zhang, G. (2024) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2024.3353263 · Cited by 62
“We solve the following visual-inertial PnP to get an intial estimation to s i : arg min A typical sliding-window based system, like OKVIS [21] or VINS-Mono [34], will conditionally mark the new frame as a keyframe.”
MethodPerception-aware Path Planning for UAVs using Semantic Segmentation
Bartolomei, L., Teixeira, L., & Chli, M. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341347 · Cited by 85
“We employ the keyframe-based VIO system VINS-Mono [22] to estimate the 6-Degrees-of-Freedom (DoF) pose of the camera using the stream of grayscale images and IMU measurements.”
MethodGraph-Based Adaptive Fusion of GNSS and VIO Under Intermittent GNSS-Degraded Environment
Gong, Z., Liu, P., Wen, F., Ying, R., Ji, X., Miao, R., & Xue, W. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3039640 · Cited by 68
“…odometry (VIO), and vision-based simultaneous localization and mapping (V-SLAM) systems under global navigation satellite system (GNSS)-denied environments have been extensively studied, which have produced fruitful outcomes, such as ORB-SLAM [1], [2], OKVIS [3], IVPR [4], and VINS-mono [5].”
BackgroundMethodInfluentialThe MADMAX data set for visual‐inertial rover navigation on Mars
Meyer, L., Smíšek, M., Fontan, A., Maza, L. O., Medina, D., Schuster, M. J., Steidle, F., Vayugundla, M., Muller, M. G., Rebele, B., et al. (2021) · Journal of Field Robotics · doi:10.1002/rob.22016 · Cited by 65
“As we mention above, we test MADMAX with two different state‐of‐the‐art SLAM baselines: ORB‐SLAM2 (Mur‐Artal & Tardós, 2017) and VINS‐Mono (Qin et al., 2018).”
MethodInfluentialXiong, L., Kang, R., Zhao, J., Zhang, P., Xu, M., Ju, R., Ye, C., & Feng, T. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3107873 · Cited by 68
Continuous-Time Vs. Discrete-Time Vision-Based SLAM: A Comparative Study
Cioffi, G., Cieslewski, T., & Scaramuzza, D. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3143303 · Cited by 75
“In the discrete-time formulation, the trajectory is represented by the body poses at the rate of the camera: The camera-IMU time offset is estimated using the method proposed in [3], which is integrated in VINS-Mono [23], a state-of-the-art visual-inertial odometry pipeline.”
MethodLi, J., Wu, W., Yang, B., Zou, X., Yang, Y., Zhao, X., & Dong, Z. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3275307 · Cited by 91
“IMU preintegration [70] is widely used to integrate IMU measurements in factor graph optimization, to maintain relative motion accuracy for a short period of time.”
MethodVisual-Inertial Localization With Prior LiDAR Map Constraints
Zuo, X., Geneva, P., Yang, Y., Ye, W., Liu, Y., & Huang, G. P. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2927123 · Cited by 77
“Over the past decades, a variety of sensors have been employed for localization in different environments [1]–[6].”
BackgroundMethodMinoda, K., Schilling, F., Wüest, V., Floreano, D., & Yairi, T. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3058073 · Cited by 75
“To this end, we develop the method VINS-Mask, which is based on VINS-Mono [4].”
MethodA GPS-aided Omnidirectional Visual-Inertial State Estimator in Ubiquitous Environments
Yu, Y., Gao, W., Liu, C., Shen, S., & Liu, M. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8968519 · Cited by 67
“(4) where [rp,Hp] is the prior information [2].”
BackgroundMethodSelf-citationTP-TIO: A Robust Thermal-Inertial Odometry with Deep ThermalPoint
Zhao, S., Wang, P., Zhang, H., Fang, Z., & Scherer, S. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341716 · Cited by 74
“• For the state estimation, we mainly compare the accuracy and robustness of TP-TIO with other state-of-theart odometry methods such as VINS [14], OKVIS [18], R-VIO [19] and DSO [20].”
BackgroundMethodResultInfluentialKayhani, N., Zhao, W., McCabe, B. Y., & Schoellig, A. P. (2022) · Automation in Construction · doi:10.1016/j.autcon.2021.104112 · Cited by 67
“Visual and visual-inertial odometry (VO [32] and VIO [33,34]) techniques are the most popular methods for local pose estimation in aerial robots [16].”
BackgroundMethodAutomatic Calibration of Multiple 3D LiDARs in Urban Environments
Jiao, J.-H., Yu, Y., Liao, Q., Ye, H.-Y., Fan, R., & Liu, M. (2019) · IEEE/RJS International Conference on Intelligent RObots and Systems · doi:10.1109/iros40897.2019.8967797 · Cited by 57
“mporal offset between sensors. Additionally, several state-of-the-art visual-inertial navigation systems adopted the motion-based approaches to achieve online calibration of camera-IMU transformation [22]. Although the motion-based calibration has been extensively developed, the accuracy of the results is easily affec…”
MethodLi, Q., Nevalainen, P., Queralta, J. P., Heikkonen, J., & Westerlund, T. (2020) · Remote Sensing · doi:10.3390/rs12111870 · Cited by 71
Enhancing navigation performance through visual-inertial odometry in GNSS-degraded environment
Liao, J., Li, X., Wang, X., Li, S., & Wang, H. (2021) · GPS Solutions · doi:10.1007/s10291-020-01056-0 · Cited by 62
Research on Multi-Sensor Fusion SLAM Algorithm Based on Improved Gmapping
Tian, C., Liu, H., Liu, Z., Li, H., & Wang, Y. (2023) · IEEE Access · doi:10.1109/access.2023.3243633 · Cited by 65
“Hong Kong University of Science and Technology Shen Shaojie et al. [14] proposed an algorithm based on the fusion of monocular vision and IMU Visual-Inertial System Monocular (IMU - VINS-Mono).”
Method
Metadata
- Authors
- Qin, T., Li, P., & Shen, S.
- Affiliations
- Hong Kong University of Science and Technology (HK)
- Venue
- IEEE Transactions on Robotics
- Published
- 2018-08
- Volume
- 34
- Issue
- 4
- Pages
- 1004-1020
- ISSN
- 1552-3098 (ISSN-L)
- 1941-0468
- OpenAlex
- W2745859992
- Cited by
- 4,669
- References
- 53
- Citations by year
- 2026: 575 · 2025: 856 · 2024: 789 · 2023: 729 · 2022: 571 · 2021: 469 · 2020: 357 · 2019: 248 · 2018: 73
- FWCI
- 233.33
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- IEEE Transactions on Robotics (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Hong Kong Research Grants Council Early Career Scheme (26201616)
- HKUST Proof-of-Concept (PCF.009.16/17)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Engineering › Aerospace Engineering › Inertial Sensor and Navigation
- Physical Sciences › Engineering › Aerospace Engineering › Aerospace Engineering and Control Systems
- Keywords
- loop detection, relocalization, pose graph optimization, autonomous flight
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.