On-Manifold Preintegration for Real-Time Visual--Inertial Odometry
Open accessThe DOI matches its Crossref record; 83 of the 83 references it cites were found in a registry or an index.
Abstract
Current approaches for visual-inertial odometry (VIO) are able to attain highly accurate state estimation via nonlinear optimization. However, real-time optimization quickly becomes infeasible as the trajectory grows over time; this problem is further emphasized by the fact that inertial measurements come at high rate, hence, leading to the fast growth of the number of variables in the optimization. In this paper, we address this issue by preintegrating inertial measurements between selected keyframes into single relative motion constraints. Our first contribution is a preintegration theory that properly addresses the manifold structure of the rotation group. We formally discuss the generative measurement model as well as the nature of the rotation noise and derive the expression for the maximum a posteriori state estimator. Our theoretical development enables the computation of all necessary Jacobians for the optimization and a posteriori bias correction in analytic form. The second contribution is to show that the preintegrated inertial measurement unit model can be seamlessly integrated into a visual-inertial pipeline under the unifying framework of factor graphs. This enables the application of incremental-smoothing algorithms and the use of a structureless model for visual measurements, which avoids optimizing over the 3-D points, further accelerating the computation. We perform an extensive evaluation of our monocular VIO pipeline on real and simulated datasets. The results confirm that our modeling effort leads to an accurate state estimation in real time, outperforming state-of-the-art approaches.
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.
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 2×
“We use the relative error metrics proposed in [68] to obtain error statistics.”
MethodEstimation with Applications to Tracking and Navigation: Theory, Algorithms and Software
Bar‐Shalom, Y., Li, X. ‐., & Kirubarajan, T. (2002) · doi:10.1002/0471221279 · Cited by 6,782
A Mathematical Introduction to Robotic Manipulation
Murray, R. M., Li, Z., & Sastry, S. S. (2017) · CRC Press eBooks · doi:10.1201/9781315136370 · Cited by 6,694 · cited 1×
“For this purpose, we introduce the following kinematic model [49], [53]:”
MethodFactor graphs and the sum-product algorithm
Kschischang, F. R., Frey, B. J., & Loeliger, H.-A. (2001) · IEEE Transactions on Information Theory · doi:10.1109/18.910572 · Cited by 6,486 · cited 1×
“This naturally leads to the well known factor graph representation, a class of bipartite graphical models that can be used to represent such factored densities [51], [52].”
BackgroundParallel 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 · cited 1×
“A breakthrough has been the development of incremental smoothing techniques (iSAM [36], iSAM2 [3]), which leverage the expressiveness of factor graphs to maintain sparsity and to identify and update only the typically small subset of variables affected by a new measurement.”
MethodMonoSLAM: Real-time single camera SLAM
Davison, A. J., Reid, I. D., Molton, N. D., & Stasse, O. (2007) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2007.1049 · Cited by 3,959 · cited 1×
“The complexity of the EKF grows quadratically in the number of estimated landmarks, therefore, a small number of landmarks (in the order of 20) are typically tracked to allow real-time operation [10–12].”
BackgroundMatrix Analysis and Applied Linear Algebra
Meyer, C. (2000) · doi:10.1137/1.9780898719512 · Cited by 3,565 · cited 1×
“Any basis El ∈ R2nl×2nl−3 of the null space of El satisfies the following relation [71]:”
BackgroundMatrix analysis and applied linear algebra
(2001) · Choice Reviews Online · doi:10.5860/choice.38-5065 · Cited by 3,300
Stochastic models, estimation, and control
(1982) · Mathematics in Science and Engineering · doi:10.1016/s0076-5392(08)x6186-7 · Cited by 3,229
Cited by
Items 901–1,000 of 1,000
Exploring the Accuracy Potential of IMU Preintegration in Factor Graph OptimizationPreprint
Tang, H., Niu, X., Zhang, T., Fan, J. F., & Liu, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.03010 · Cited by 16
“The noise covariance strongly influences the MAP estimator, because the inverse noise covariance (information matrix) is used to weight the factor in the FGO [11], [13].”
Fast-Lio: A Fast, Robust Lidar-Inertial Odometry Package by Tightly-Coupled Iterated Kalman FilterPreprint
Xu, W., & Zhang, F. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4675587 · Cited by 4
Smoothing algorithms for navigation, localisation and mapping based on high-grade inertial sensors
Chauchat, P. (2020) · theses.fr (ABES) · doi:10.70675/da76b43cz1bb8z4971zadcfz1d6a52390d7c · Cited by 6
“From a computational point of view, SC-BIFM is promising, as its complexity is reasonably low since only matrices of limited size need to be inverted, as opposed to the robust batch solver.”
BackgroundMethodInfluentialDynamic Resource-Aware Corner Detection for Bio-Inspired Vision Sensors
Mohamed, S. A. S., Yasin, J. N., Haghbayan, M.-H., Miele, A., Heikkonen, J., Tenhunen, H., & Plosila, J. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412314 · Cited by 16
“For example, most of the vSLAM approaches in the state-of-the-art [4]–[7] are based on analyzing images captured by traditional framebased cameras to obtain the translation and rotation between two consecutive images.”
BackgroundLIVS: a tightly coupled LiDAR, IMU, and camera-based SLAM system under scene degradation
Cheng, G., Li, P., Li, Q., Wang, D., Li, Z., & Wang, Z. (2024) · Physica Scripta · doi:10.1088/1402-4896/ad859f · Cited by 3
Xiao, Y., Xu, W., Li, B., Zhang, H., Xu, B., & Zhou, W. (2024) · Sensors · doi:10.3390/s24227360 · Cited by 3
Visual and LiDAR SLAM Performance in Campus Scenes
Zhou, T., Fang, R., Wang, F. L., Wang, G., Wu, L., Su, K., Xu, L., Pu, H., & Luo, J. T. (2024) · International Conference on Service Robotics (ICoSR) · doi:10.1109/icosr63848.2024.00027 · Cited by 3
SLAM Method of Mine Inspection Robot based on Stereo Vision and IMU
Liu, H., Jiana, P., Zheng, F., Han, K., & Zhang, A. (2024) · China Automation Congress (CAC) · doi:10.1109/cac63892.2024.10864764 · Cited by 3
Robust Multi-Sensor Fusion Positioning Based on GNSS/IMU Using Factor Graph Optimization
Ahmadi, E., Elsanhoury, M., Selvan, K., Välisuo, P. O., & Kuusniemi, H. (2025) · IEEE/ION Position Location and Navigation Symposium · doi:10.1109/plans61210.2025.11028340 · Cited by 3
“To meet real-time navigation requirements, the IMU pre-integration technique introduced in [22], [23] is used.”
Tran, Q. D., Hoang, Q. H., & Kim, G.-W. (2025) · IEEE Access · doi:10.1109/access.2025.3582252 · Cited by 4
“…: Finally, the pose of the system at time t + (cid:49) t can be updated as follows : where ˆ R WB = R W The relative motion of the IMU between two consecutive nodes, can be determined using the method outlined based on [36], which calculates the position, velocity, and orientation as follows :”
Distributed collaborative navigation for UAV formation based on factor graph optimization
Guo, P., Zhang, R., & Zhang, J. (2025) · Measurement · doi:10.1016/j.measurement.2025.118732 · Cited by 6
Shu, Z., Bei, S., Li, L., & Chen, Z. (2025) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2025.3616838 · Cited by 5
“Moreover, when estimating high-frequency GT trajectories, preintegration-based approaches are prone to degeneracy [16,17], which further restricts their effectiveness.”
Tightly Coupled Visual–Inertial SLAM With Dual-Modal Confidence Perception
Zhao, J., Zhang, X., He, K., Qi, Z., & Yang, Y. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3621699 · Cited by 3
“According to the pre-integration theory [16], the update formulas for the corresponding mean acceleration A i and mean angular velocity ω i of image i are given by where t imu denotes the IMU pre-integration time and A bai and ω b ω i correspond to the bias-inclusive accelerometer and gyroscope…”
LEVIO: Lightweight Embedded Visual Inertial Odometry for Resource-Constrained Devices
Kühne, J., Vogt, C., Magno, M., & Luca Benini, C. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3644788 · Cited by 3
“2) IMU Pre-Integration: To account for both effects of IMU measurements arriving at different rates than image frames and keyframes being selected at irregular intervals, we employ IMU pre-integration [34] to obtain only one IMU constraint between two keyframes.”
IMU-Preintegrated Radar Factors for Asynchronous Radar-LiDAR-Inertial SLAM
Hatleskog, J., Nissov, M., & Alexis, K. (2025) · International Conference on Advanced Robotics (ICAR) · doi:10.1109/icar65334.2025.11338704 · Cited by 3
“We create state nodes in the factor graph only at the timestamps of incoming LiDAR measurements and connect them with preintegrated IMU measurements, following [9].”
InfluentialA Plug-and-Play Learning-Based IMU Bias Factor for Robust Visual–Inertial Odometry
Yang, Y., Wang, K., Zhang, J., Tan, Z., Wang, X., Shen, H., & Hu, D. (2026) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2026.3676182 · Cited by 6
“2) Preintegration: We follow the discrete expressions in [33] and express the displacement, velocity, and rotational preintegration terms as (2) Fig.”
Aerial Robotic System for Complete Bridge Inspections
Viguria, A., Caballero, R., Petrus, A. L., Pérez-Grau, F. J., & Trujillo, M. A. (2022) · Sustainable civil infrastructures · doi:10.1007/978-3-030-79801-7_12 · Cited by 3
Multi-Sensor Fusion System Based on Factor Graph Optimization
Cheng, X., Peng, X., Wu, D., & Zhang, X. (2023) · Chinese Control Conference (CCC) · doi:10.23919/ccc58697.2023.10240028 · Cited by 3
“Therefore, IMU pre-integration [7] is to calculate the relative motion and update the position and orientation to the coordinate of the previous moment to reduce the calculation amount.”
BackgroundThe First Experimental Study of Monocular Visual Localization for Train using Deep Learning Method
Wang, Y., Xie, J., Zhang, J., Li, S., Fan, K. B., & Zhang, M. (2023) · IEEE International Conference on Cloud Computing and Intelligence Systems (CCIS) · doi:10.1109/ccis59572.2023.10262972 · Cited by 3
Robot Localization and Reconstruction based on 3D Point Cloud
Chi, P., Wang, Z., Liao, H., Wu, X., Tian, J., & Zhang, Q. (2023) · IEEE RO-MAN · doi:10.1109/ro-man57019.2023.10309593 · Cited by 3
“Since the frequency of the IMU is much higher than that of point cloud data and the integral operation takes a lot of time, the pre-integration method proposed in [27] is used to reduce the time.”
MethodZhou, W., Zhang, H., Fan, C., Hu, H., He, X., Zhang, L., & Hu, X. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3433375 · Cited by 3
“Further, the theory of integral increment between key frames is perfected, and the error transfer mode of inertial integration is analyzed mathematically [45].”
BackgroundA Comprehensive Study of Systems Challenges in Visual Simultaneous Localization and Mapping Systems
Semenova, S., Ko, S. Y., Liu, Y. D., Ziarek, L., & Dantu, K. (2024) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3677317 · Cited by 8
Visual-Inertial Odometry Using Optical Flow from Deep Learning
Kang, J. M., Sjanic, Z., & Hendeby, G. (2024) · International Conference on Information Fusion (FUSION) · doi:10.23919/fusion59988.2024.10706322 · Cited by 4
“The last mentioned, VINS -Mono, employs an IMU pre-integration method, [16], which integrates the IMU parameters between camera frames, as well as modularized pipeline for the visual-inertial estimator.”
MethodRobot Localization Method Based on Multi-Sensor Fusion in Low-Light Environment
Wang, M., Lian, Z., Andrés, M. A. N., Wang, P., Tian, Y., Yue, Z., & Gu, L. (2024) · Electronics · doi:10.3390/electronics13224346 · Cited by 4
Liu, J., Qi, Y., Yuan, G., Liu, L., & Li, Y. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad99f5 · Cited by 3
“It requires only the integration of IMU motion increments between two frames and the previous integration, addressing the issue in which repeated integrations severely constrain computational resources and real-time performance [28].”
A Biomimetic Pose Estimation and Target Perception Strategy for Transmission Line Maintenance UAVs
Zhuo, H., Yang, Z. X., Zhang, C., Xu, N., Xue, B., Zhu, Z., & Xie, Y. (2024) · Biomimetics · doi:10.3390/biomimetics9120745 · Cited by 4
“Foster et al. [30] built a complete theoretical system through applying IMU pre-integration to Lie algebra.”
Self-TIO: Thermal-Inertial Odometry via Self-Supervised 16-Bit Feature Extractor and Tracker
Lee, J., Ando, T., Shinozaki, M., Kitajima, T., An, Q., & Yamashita, A. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3518303 · Cited by 4
Liu, B., Liu, H., Xing, Y., Gong, W., Yang, S., Yang, H. W., Pan, K., Li, Y., Hou, Y., & Jia, S. (2025) · Forests · doi:10.3390/f16020335 · Cited by 4
General measurement model of distributed accelerometers for inertial navigation
Chang, G., Ye, Y., Gong, S., & Yuan, Y. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/adc9cf · Cited by 3
“…be readily fused optimally with the inertial data in the proposed model, e.g. satellite-navigation data [2], acoustic-positioning data [31], DVL data [32], visual odometry data [33], attitude data from star trackers or magnetometers [16], terrain data for matching [34], vehicle dynamics model [35].”
What’s Wrong with the Absolute Trajectory Error?
Lee, S. H., & Civera, J. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-91569-7_8 · Cited by 9
An Approach Based on BiLSTM-Transformer for Marine Inertial Navigation
Du, N., Yu, H., Sheng, H., He, Q., Tu, J., & Han, Z. (2025) · International Symposium on Computer Applications and Information Systems (ISCAIS) · doi:10.1109/iscait64916.2025.11010712 · Cited by 3
“The traditional method directly recursively integrates the discrete IMU sample data, resulting in a large number of repetitive operations as the entire integration path needs to be recalculated at each GNSS update [9] .”
Wu, Y., Li, Y., & Shi, B. (2025) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-025-02300-w · Cited by 3
“2 [49]: where Exp denotes the exponential map: Exp : R 3 → SO ( 3 ) . ω j denotes the gyroscope raw measurement data, b g is the gyroscope bias calculated from the IMU initialization before vision loss, and n jg is the additive white noise of the gyroscope measurement.”
HiMAR: a hierarchical context-aware mobile AR assembly for large-scale cable products
Fang, W. H., Zhang, T., Chen, L., Cao, Y., Hu, H., & Bi, J. (2025) · International Journal of Production Research · doi:10.1080/00207543.2025.2568740 · Cited by 3
“In this study, the well-known IMU pre-integration algorithm (Forster et al. 2017) is applied to construct the inertial factor, and the Gaussian white noise is assumed to be acceleration and gyroscope measurements.”
3-D LiDAR-Aided GNSS NLOS Mitigation for Reliable GNSS-RTK Positioning in Urban Canyons
Liu, X., Wen, W., Huang, F., Gao, H., Wang, Y., & Hsu, L. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3629836 · Cited by 9
“readers can refer to [3, 39] for the detailed formulation of the inertial factors.”
Mo, S., Zhou, Z., Du, S., Xiang, C., Kuang, C., & Wu, J. (2018) · Lecture notes in electrical engineering · doi:10.1007/978-981-13-0029-5_54 · Cited by 3
An improved method for odometry estimation based on EKF and Temporal Convolutional Network
Leng, X., Piao, S., Wang, S., Chang, L., & Zhu, Z. H. (2020) · Physical Communication · doi:10.1016/j.phycom.2020.101178 · Cited by 3
Optical Flow Monocular Visual-Inertial Odometry with Online Photometric Calibration
Liu, Q., Wang, Z., & Wang, H. (2021) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1828/1/012163 · Cited by 3
“We adopt the empiric model of response (EMoR) proposed by Grossberg and Nayar [8] to model the response function.”
MethodInitialization Improvement and Map Reuse Based on ORBSLAM3
Feng, L. (2021) · International Conference on Artificial Intelligence and Information Systems (ICAIIS) · doi:10.1145/3469213.3470677 · Cited by 3
“The SLAM system is based on the theory of IMU preintegration on manifold [7], [8] to achieve a visual-inertial tightly coupled Bundle Adjustment algorithm.”
MethodInformation sparsification for visual-inertial odometry by manipulating Bayes tree
Zhang, X., Yuan, W., Zheng, L., Zhan, X., Ren, Q., & Lin, B. (2021) · Aerospace Science and Technology · doi:10.1016/j.ast.2021.107065 · Cited by 3
PLMCVIO: Point-Line based Multi-Camera Visual Inertial Odometry
Wang, F., Zhang, C., Zhang, G., Liu, Y., Xia, Y., & Yang, X. (2021) · IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems · doi:10.1109/cyber53097.2021.9588162 · Cited by 3
Zhang, X., Yuan, W., Zhan, X., Chi, C., Ren, Q., Gong, W., & Lin, B. (2021) · Aerospace Science and Technology · doi:10.1016/j.ast.2021.107250 · Cited by 3
Jongeneel, M. J., & Saccon, A. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981409 · Cited by 5
“Besides the potential issue emerging from the well-known problem of running into representation singularity [12], [13], the approach is somehow affected by the problem that the selected parametrization (Euler angles, Cardan angles, Roll-Pitch-Yaw angles, etc.) will provide non-uniform filtering of…”
BackgroundContinuous-Time Trajectory Estimation for Differentially Flat Systems
Johnson, J. C., Mangelson, J. G., & Beard, R. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3224364 · Cited by 5
“Methods such as inertial measurement unit (IMU) preintegration [3] can be introduced to allow for using such sensors, but these methods may not be trivial and must be carefully tailored to each individual sensor type.”
BackgroundDistributed Moving Horizon Fusion Estimation for Nonlinear Constrained Uncertain Systems
Wang, S., & Xue, B. (2023) · Mathematics · doi:10.3390/math11061507 · Cited by 3
Hejun, 韦. W., Enyong, 许. X., Bing, 韩. H., Yanmei, 蒙. M., Jin, 韦. W., & Zhengqiang, 李. L. (2023) · Laser & Optoelectronics Progress · doi:10.3788/lop213074 · Cited by 3
Visual-Inertial-Laser SLAM Based on ORB-SLAM3
Cao, M., Zhang, J., & Chen, W. (2023) · Unmanned Systems · doi:10.1142/s2301385024500262 · Cited by 4
Tsao, S.-H., & Jan, S. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3273510 · Cited by 4
“To avoid reintegration, IMU preintegration [8], [9] defines a factor independent from the state estimate using only inertial measurements between two keyframes.”
MethodA tightly-coupled method of lidar-inertial based on complementary filtering
Liu, J., Zheng, J., Jia, X., Li, T., & Zhang, W. (2023) · Measurement Science and Technology · doi:10.1088/1361-6501/ace20e · Cited by 3
“The Euler angle increment between two frames of point cloud data is calculated by IMU pre-integration technique [20].”
Yu, C.-H., Chao, Z., Xie, H., Hua, Y., & Wu, W. (2023) · Robotics · doi:10.3390/robotics13010002 · Cited by 6
LIO-DOR: A Robust LiDAR Inertial Odometry with Real-time Dynamic Object Removal
Mao, L., Gao, W., Chen, H., & Zhang, S. (2023) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio58561.2023.10355038 · Cited by 3
“When the IMU measurement at time t + ∆ t is received by the system, the IMU preintegration based on the pose at previous time t can be performed as where R ∆ t is the rotation matrix between the two frames [9], and p ∆ t is the translation vector between the two frames.”
MethodBao, M., Dai, K., Wang, K., Fan, Z., Xu, R., Li, R., & Zhou, H. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3385822 · Cited by 3
“In the framework of an EKF, numerical integration [54], [55] is used to predict the prior robot state ˇ ξ i , including position ˇ p i , velocity ˇ v i , and attitude ˇ q i .”
MethodLee, J., Kurisu, M., & Kuriyama, K. (2024) · Journal of Field Robotics · doi:10.1002/rob.22334 · Cited by 3
“LIO ‐ SAM achieved high accuracy and frequency of pose estimation compared to the LOAM by effectively fusing high ‐ frequency IMU measurements with LOAM ‐ based mapping using IMU preintegration (Forster et al., 2016) and pose graph (Kaess et al., 2012).”
MethodRobust Factor Graph Optimization Integrated Navigation Based on Improved Chi-Square Test
Zhou, Y., Chen, X., & Ge, M. (2023) · International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD) · doi:10.1109/icsmd60522.2023.10491039 · Cited by 3
“The recursive integration representation of the INS state over the integration time interval 1 [ , ] k k t t − can be expressed as[11][12]: Navigation frame n denotes North-East-Up (ENU), body frame b denotes Right-Forward-Up (RFU), and k is time k t .”
BackgroundYin, J., Zhuang, Y., Yan, F., Liu, Y., & Zhang, H. (2024) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2024.3389012 · Cited by 4
“The inertial residual r I ( i-1 ,i ) between frame i-1 and i is the same as the one in [23], which is defined as where Log: SO (3) → R 3 ∗ 1 maps the rotation matrix from the Lie group to the vector space.”
BackgroundBo, 张. Z., Yinlong, 张. Z., Wei, 梁. L., Xin, 王. W., & Yutuo, 杨. Y. (2024) · Acta Optica Sinica · doi:10.3788/aos231613 · Cited by 3
Visual-Inertial Odometry for Metric-Scale Mapping of Underwater Caves
Lago, A., Neves, N., Ventura, R., & Gaspar, J. A. (2024) · IEEE International Conference on Autonomous Robot Systems and Competitions · doi:10.1109/icarsc61747.2024.10535924 · Cited by 6
“As detailed in [8] and [17], the UKF implementation implies linearization and discretization of the system’s dynamics, (2) and (3).”
BackgroundDynaVIG: Monocular Vision/INS/GNSS Integrated Ego and Object Localization in Dynamic Scenes
Jin, R., Wang, Y., Mei, T., Niu, X., Liu, J., Hu, Y., He, J., & Hu, J. (2024) · International Conference on Control and Automation · doi:10.1109/icca62789.2024.10591957 · Cited by 3
“We use the method in [16] to compute IMU pre-integration between two epoches t i and t j : where (cid:101) ω k , (cid:101) a k are the angle velocity and acceleration measurements, ∆ t is the IMU time interval, ∆ R ij , ∆ v ij , ∆ p ij represents the preintegrated rotation, velocity and translation…”
MethodVisual Inertial Odometry using Focal Plane Binary Features (BIT-VIO)
Lisondra, M., Kim, J., Murai, R., Zareinia, K., & Sajad Saeedi, G. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610838 · Cited by 5
“For example, one can use filtering [2], [30] approaches, or using non-linear optimization, perform fixed-lag smoothing [23], [36] or even full smoothing [12].”
MethodEnhancing Visual Inertial SLAM with Magnetic Measurements
Joshi, B., & Rekleitis, I. M. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10611341 · Cited by 6
“We leverage the IMU preintegration algorithm from [3], [4] to efficiently compute magnetometer residuals for all measurements.”
MethodInfluentialCamera Motion Estimation from RGB-D-Inertial Scene Flow
Cerezo, S., & Civera, J. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00089 · Cited by 5
“For the computation of the above matrices, the preinte-grated expressions in [4] were used.”
BackgroundMethodInfluentialDGRO: Doppler Velocity and Gyroscope-Aided Radar Odometry
Guo, C., Wei, B., Lan, B., Liang, L., & Liu, H. (2024) · Sensors · doi:10.3390/s24206559 · Cited by 3
“Thus, when other sensors fail, integrating IMU data can significantly enhance positioning accuracy and improve the overall performance and robustness of the SLAM system [17].”
BackgroundVisual-Inertial Localization Leveraging Skylight Polarization Pattern Constraints
Wan, Z., Fu, P., Wang, K., & Zhao, K. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3495375 · Cited by 4
Visual Inertial SLAM Based on Spatiotemporal Consistency Optimization in Diverse Environments
Pu, H., Luo, J., Wang, G., Huang, T., Wu, L., Xiao, D., Liu, H., & Luo, J. (2024) · Journal of Field Robotics · doi:10.1002/rob.22487 · Cited by 4
Asynchronous Event-Inertial Odometry using a Unified Gaussian Process Regression Framework
Li, X., Wang, Z., Liu, Z., Zhang, Y., Zhang, F., Yao, X., & Huang, P. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802357 · Cited by 5
“A complete form of IMU preintegration is provided in [2].”
Lei, J., Liu, J., Zhang, W., Li, M., & Hyyppa, J. M. (2025) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2025.01.019 · Cited by 3
A Novel Error Model Based on Quaternion of Body Frame of SE(3) for Inertial-Integrated Navigation
Zhu, T., Li, A., Li, K., Qian, L., Li, W., & Chen, Y. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3542865 · Cited by 3
“The experts in the relevant field described the dynamic model in a flow form and overall improved the estimation accuracy [4].”
Arbabmir, M., & Ebrahimi, M. (2019) · IET Image Processing · doi:10.1049/iet-ipr.2019.0007 · Cited by 3
“Furthermore, the optimisation base algorithms [10–13] have been developed for high accuracy application.”
MethodA progressive online external parameter calibration and initialization method for stereo-IMU system
Zhai, Y., Feng, H., & Fu, Y. (2021) · Industrial Robot the international journal of robotics research and application · doi:10.1108/ir-07-2021-0135 · Cited by 3
Deep Learning based-State Estimation for Holonomic Mobile Robots Using Intrinsic Sensors
Nam, D. V., & Kim, G.-W. (2021) · International Conference on Control, Automation and Systems (ICCAS) · doi:10.23919/iccas52745.2021.9650051 · Cited by 3
Monocular Visual-Inertial SLAM with Camera-IMU Extrinsic Automatic Calibration and Online Estimation
Pan, L., Tian, F., Ying, W., & She, B. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-27538-9_61 · Cited by 3
Research on Robot Positioning Technology Based on Multi Sensor
Yu, F., Yu, H., & Wei, Y. (2019) · International Conference on Computer Network, Electronic and Automation (ICCNEA) · doi:10.1109/iccnea.2019.00094 · Cited by 3
“[9] Leutenegger S,Lynen S,Bosse M,et al Keyframe-Based Visual-Inertial Odometry Using Nonlinear Optimization [J] The International Journal of Robotics Research, 2014, 34 (3):314-334 [10] Forster C, Carlone L, Dellaert F, et al.”
BackgroundMethodReal-time location estimation for indoor navigation using a visual-inertial sensor
Wang, Z., Li, X., Zhang, X., Bai, Y., & Zheng, C. (2020) · Sensor Review · doi:10.1108/sr-01-2020-0014 · Cited by 3
State Estimation with Event-Based Inputs Using Stochastic Triggers
Noack, B., Funk, C., Radtke, S., & Hanebeck, U. D. (2020) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2020.12.2491 · Cited by 3
“Also, a combination with advanced pre-integration techniques, as proposed by Forster et al. (2017), may represent a promising research direction.”
BackgroundInfluentialImproving accuracy of VI-SLAM with fish-eye camera based on biases of map points
Kang, Z., & Zou, W. (2020) · Advanced Robotics · doi:10.1080/01691864.2020.1815573 · Cited by 3
Autonomous positioning of omnidirectional mobile robot based on visual inertial navigation
Li, P., Leng, B., & Fu, H. (2020) · Chinese Control Conference (CCC) · doi:10.23919/ccc50068.2020.9189018 · Cited by 3
“Write the above specific force equation into the matrix form: The pre-integral processing of IMU is first proposed in [11] and further improved in [12].”
BackgroundImproving the Applicability of Visual Peer-to-Peer Navigation with Crowdsourcing
Dong, E., Liang, J., Wang, Z., Xu, J., Shangguan, L., Ma, Q., & Yang, Z. (2020) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads51040.2020.00034 · Cited by 3
“With the advance of preintegration methods [24], IMU observations can be optimized together with visual observations, enabling complete systems to be more accurate, among which VINS-Mono [15] is one of the state-of-the-art systems.”
MethodA Factor Graph Method for AUV Navigation in the Mobile Docking Progress
Ruan, L., Chen, S., Zhou, J., Zeng, D., & Xu, Y. (2020) · Global Oceans 2020: Singapore – U.S. Gulf Coast · doi:10.1109/ieeeconf38699.2020.9389322 · Cited by 5
“Especially, to control the number of factor in the factor graph and perform a real time optimization, we use the IMU PREINTEGRATION method introduced in [8].”
MethodInfluentialResearch on impact resistance and active–passive compliance control of rope-driven joint unit
Cao, X. Y., & Wu, W. (2021) · Journal of the Brazilian Society of Mechanical Sciences and Engineering · doi:10.1007/s40430-021-03123-0 · Cited by 3
Inertial Aided 3D LiDAR SLAM with Hybrid Geometric Primitives in Large-scale Environments
Chen, W., Zhao, H., Shen, Q., Xiong, C., Zhou, S., & Liu, Y. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561511 · Cited by 3
“Inspired by the successful applications of preintegration theorem in VIO [6] [7], Ye et al. [8] proposed a tightly-coupled LiDAR-IMU odometry algorithm by jointly optimizing the relative LiDAR factors and IMU preintegration factors in a sliding-window.”
MethodExponential stability of an attitude trajectory tracking controller utilizing unit quaternions
Koutras, L., & Doulgeri, Z. (2021) · European Control Conference (ECC) · doi:10.23919/ecc54610.2021.9654947 · Cited by 3
RLP‐VIO: Robust and lightweight plane‐based visual‐inertial odometry for augmented reality
Li, J., Zhou, X., Yang, B., Zhang, G., Wang, X., & Bao, H. (2022) · Computer Animation and Virtual Worlds · doi:10.1002/cav.2046 · Cited by 5
A Sensor Fusion Algorithm: Improving State Estimation Accuracy for a Quadruped Robot Dog
Zhao, Q., Shao, H., Yang, W., Chen, B., Feng, Z., Teng, H., & Li, Q. (2022) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio55434.2022.10011894 · Cited by 3
“[8] solved the manifold structure problem of the rotation group based on the IMU pre-integration of Forster.”
MethodVINS-Mask: A ROI-mask Feature Tracker for Monocular Visual-inertial System
Sun, J., Song, F., & Ji, L. (2022) · International Conference on Automation, Robotics and Computer Engineering (ICARCE) · doi:10.1109/icarce55724.2022.10046501 · Cited by 3
“ORBSLAM-VI [9] reuse a map with longterm, mid-term, and short-term data association, and use them in an accurate local visual-inertial Bundle Adjustment with IMU pre-integration [10, 11].”
MethodReal-time Indoor Scene Reconstruction with RGBD and Inertia InputPreprint
Zhu, Z., & Xu, F. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.03015 · Cited by 5
“…we denote N as the number of inertial measurements acquired in two consecutive camera frames, and then merge them together by pre-integration method[11] to predict camera rotation ∆ R and translation ∆ t between two consecutive camera frames: In the above three equations, the subscript ’ n ’…”
MethodContinuous Direct Sparse Visual Odometry from RGB-D ImagesPreprint
Ghaffari, M., Clark, W., Bloch, A. M., Eustice, R. M., & Grizzle, J. W. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.02266 · Cited by 3
“Another interesting idea is guiding the gradient flow of SE(3) using an Inertial Measurement Unit (IMU) [11, 15, 19] or combined contactIMU process models [18, 19].”
BackgroundMethodGeometric Structure Aided Visual Inertial LocalizationPreprint
Huang, H., Ye, H., Jiao, J., Sun, Y., & Liu, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.04173 · Cited by 5
The Unified Mathematical Framework for IMU Preintegration in Inertial-Aided Navigation SystemPreprint
Luo, Y., Liu, Y., Guo, C., & Liu, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.09100 · Cited by 3
“There two close-form expressions for the local increment based on two basic assumptions, namely piecewise constant global acceleration [2] and piecewise constant IMU measurements [4].”
MethodLong, B., Goodin, S., Kachergis, G., Marchman, V. A., Radwan, S. F., Sparks, R. Z., Xiang, V., Zhuang, C., Hsu, O., Newman, B., et al. (2022) · doi:10.31234/osf.io/238jk · Cited by 3
Bredu, J. A. (2022) · The Research Repository @ WVU (West Virginia University) · doi:10.33915/etd.11495 · Cited by 3
System Calibration of a Field Phenotyping Robot with Multiple High-Precision Profile Laser Scanners
Esser, F., Tombrink, G., Cornelißen, A., Klingbeil, L., & Kuhlmann, H. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610208 · Cited by 2
“More information on the IMU preintegration and with bias estimation gives [24]–[26].”
BackgroundEvent-based vision in laser welding: An approach for process monitoring
Dold, P. M., Nadkarni, P., Boley, M., Schorb, V., Wu, L., Steinberg, F., Burggräf, P., & Mikut, R. (2025) · Journal of Laser Applications · doi:10.2351/7.0001526 · Cited by 6
LiDAR-Aided Object Visual-Inertial Odometry Using Anchored Residual in Dynamic Scene
Wu, J., Cheng, X., Liu, F., & Tang, X. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3552765 · Cited by 2
“…and drone, visual-inertial odometry has been widely responsible for the realization of location services via properties of low-cost, complementarity and easy deployment of camera and inertial measurement unit (IMU) in autonomous driving scene and underground space parking [1], [2], [3], [4].”
IMU/Camera/GNSS/UWB Integrated Localization Method Based on Factor Graph Optimization
Yang, R., Wang, N., Xu, Z., Huang, Y., & Zhong, X. (2025) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc65474.2025.11091141 · Cited by 2
Yun, S., Shin, J., Cha, J., & Kim, A. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3601033 · Cited by 2
Liu, W., Qi, Y., Guan, T., & Liang, Y. (2025) · Journal of Civil Structural Health Monitoring · doi:10.1007/s13349-025-01005-w · Cited by 2
Teng, X., Xu, S., Guo, D., Guo, Y., Xu, P., Hu, R., & Chai, H. (2025) · Computer Networks · doi:10.1016/j.comnet.2025.111710 · Cited by 2
Modular Acoustic Graph SLAM for Underwater Monitoring With Autonomous Underwater Vehicles
Real, M., Vial, P., Pi, R., Palomeras, N., & Carreras, M. (2025) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2025.3615829 · Cited by 2
“Once a new key frame is created, the previous preintegration computation stops and builds the preintegrated inertial factor [40], which contains the pose increment between two key frames of the graph in the SE (3) group.”
Han, B., Li, T., Lv, Y., Wen, G., Wang, Z., & Shi, C. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3615276 · Cited by 3
“Referring to [38], we can compute the IMU residual r I i , i + 1 based on the preintegration term along with its corresponding covariance matrix Σ I i , i + 1 .”
Liu, S., Yang, B., Tang, Q., Li, S., Hu, W., Yuan, H., & Zhu, L. (2025) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-025-02325-1 · Cited by 2
“Building on this, Forster et al. introduced Lie groups and Lie algebras to represent rigid body rotations in 3D space within the VI-SLAM system, enhancing the real-time performance and robustness of tightly coupled nonlinear optimization methods [6].”
A Robust, Real-Time, and Drift-Reduced LiDAR–Inertial–UWB Tightly Coupled Fusion
Chen, Z., Jin, S., Bian, G., Sui, X., & Xu, A. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3630840 · Cited by 2
“In terms of external sensing sensors, SLAM can be divided into two main categories, namely, visual-SLAM and light detection and ranging (LiDAR)-SLAM [6], [7].”
Metadata
- Authors
- Forster, C., Carlone, L., Dellaert, F., & Scaramuzza, D.
- Affiliations
- Massachusetts Institute of Technology (US)
- Georgia Institute of Technology (US)
- University of Zurich (CH)
- Venue
- IEEE Transactions on Robotics
- Published
- 2017-02
- Volume
- 33
- Issue
- 1
- Pages
- 1-21
- ISSN
- 1552-3098 (ISSN-L)
- 1941-0468
- OpenAlex
- W2482726005
- Cited by
- 1,502
- References
- 83
- Citations by year
- 2026: 156 · 2025: 179 · 2024: 234 · 2023: 193 · 2022: 180 · 2021: 174 · 2020: 134 · 2019: 129 · 2018: 88 · 2017: 32
- FWCI
- 102.35
- Type
- journal article
- Language
- en
- Open access
- green
- www.zora.uzh.ch
- Licence
- Locations
- IEEE Transactions on Robotics (publishedVersion)
- Zurich Open Repository and Archive (University of Zurich) (acceptedVersion, other-oa)
- arXiv (Cornell University) (submittedVersion)
- Infoscience (Ecole Polytechnique Fédérale de Lausanne) (submittedVersion)
- Infoscience (Ecole Polytechnique Fédérale de Lausanne) (submittedVersion, cc-by-nc-nd)
- Zurich Open Repository and Archive (University of Zurich) (submittedVersion, other-oa)
- arXiv (Cornell University)
- Zurich Open Repository and Archive (University of Zurich)
- Funders
- Swiss National Foundation (200021-143607)
- National Center of Competence in Research Robotics (NCCR)
- University of Zurich
- National Science Foundation (11115678)
- United States Department of Agriculture (USDA)
- National Institute of Food and Agriculture (GEOW-2014-09160)
- 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 › Computer Science › Artificial Intelligence › Target Tracking and Data Fusion in Sensor Networks
- Keywords
- visual-inertial odometry, inertial measurement unit, factor graphs, maximum a posteriori estimation, bias correction, real-time state estimation, keyframe selection
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.