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 601–700 of 1,000
Stereo Visual-Inertial Odometry With Online Initialization and Extrinsic Self-Calibration
Yin, H., Liu, P. X., & Zheng, M. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3282674 · Cited by 11
“Therefore, αbi b j , βbi b j , and qbi b j are the function of IMU biases and they can be adjusted by first-order Taylor expansion approximations [1]”
BackgroundMethodInfluentialCt-LVI: A Framework Toward Continuous-Time Laser-Visual-Inertial Odometry and Mapping
Wang, Z., Zhang, L., Zhao, S., & Zhou, Y. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3335989 · Cited by 14
“…point clouds and IMU readings in the time window, we first resort to Normal Distribution Transform (NDT) [45] to obtain the relative motion of the multi-frame point clouds in a short-period time window, and at the same time obtain the IMU pre-integration values resorting to Forster’s theory [46].”
MethodDI-EME: Deep Inertial Ego-Motion Estimation for Autonomous Underwater Vehicle
Li, Z., Yu, H., Yang, W., Zhang, Y., Li, Y., & Xiao, H. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3386354 · Cited by 10
“To address the drawback, the relative motion increments are introduced [39].”
BackgroundDirect visual-inertial odometry with semi-dense mapping
Xu, W., Choi, D., & Wang, G. (2018) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2018.02.002 · Cited by 11
“We take the preintegrated measurements ∆p kG ∆v kG ∆R kG as: According to infinitesimal increment in so(3) with right hand-multiplication [8,2] exp ([ θ Then, we take the same fist-order approximation for the logarithm introduced in”
Detection and Resolution of Motion Conflict in Visual Inertial Odometry
Babu, B. P. W., Cyganski, D., Duckworth, J., & Kim, S. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8460870 · Cited by 11
“The linear acceleration and angular velocity obtained from IMU sensors are fused using an Extended Kalman Filter or nonlinear optimization in loosely coupled [10] or tightly coupled integration [11], [12], [13].”
MethodTsao, S.-H., & Jan, S. (2018) · IEEE Sensors Journal · doi:10.1109/jsen.2018.2886764 · Cited by 10
Wu, J., Liu, M., & Qi, Y. (2019) · IEEE Sensors Journal · doi:10.1109/jsen.2019.2924668 · Cited by 12
Covariance Pre-Integration for Delayed Measurements in Multi-Sensor Fusion
Allak, E., Jung, R., & Weiß, S. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8967886 · Cited by 14
VI-SLAM for Subterranean Environments
Kramer, A., Kasper, M., & Heckman, C. (2021) · Springer proceedings in advanced robotics · doi:10.1007/978-981-15-9460-1_12 · Cited by 12
“These IMU readings are preintegrated as in [4].”
Monocular Visual Inertial Direct SLAM with Robust Scale Estimation for Ground Robots/Vehicles
Sahoo, B., Biglarbegian, M., & Melek, W. (2021) · Robotics · doi:10.3390/robotics10010023 · Cited by 8
“[28], a method is described for unifying multiple IMU measurements into a single factor and sparse landmark features in a structureless approach in a factor-graph [29] framework.”
BackgroundMethodAerial Robotic Solution for Detailed Inspection of Viaducts
Caballero, R., Parra, J., Trujillo, M. A., Pérez-Grau, F. J., Viguria, A., & Ollero, A. (2021) · Applied Sciences · doi:10.3390/app11188404 · Cited by 10
“The IMU pre-integration is based on [37].”
Semi-Direct Monocular Visual-Inertial Odometry Using Point and Line Features for IoV
Jiang, N., Huang, D., Chen, J., Wen, J., Zhang, H., & Chen, H. (2021) · ACM Transactions on Internet Technology · doi:10.1145/3432248 · Cited by 12
“The IMU-related Jacobians are described in [5]:”
BackgroundMethodInfluentialVisual-Laser-Inertial SLAM Using a Compact 3D Scanner for Confined Space
Cheng, D., Shi, H., Xu, A., Schwerin, M., Crivella, M., Lü, L., & Choset, H. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561860 · Cited by 11
“We perform preintegration following [29], [23].”
BackgroundXiao, K., Yu, W., Liu, W., Qu, F. J., & Ma, Z. (2022) · Applied Sciences · doi:10.3390/app12030939 · Cited by 12
“Due to space limitations, we invite readers to refer to the descriptions in [26,34] to understand the detailed derivation of IMU pre-integration.”
BackgroundImproving robustness of line features for VIO in dynamic scene
Wu, J., Xiong, J. Q., & Guo, H. (2022) · Measurement Science and Technology · doi:10.1088/1361-6501/ac547f · Cited by 11
“Visual-inertial odometry (VIO) [1–3] is developing rapidly on the strength of the fusion of information from visual simultaneous localization and mapping (SLAM) [4–6] and inertial measurement, and it is widely used for automated driving, augmented reality and unmanned aerial vehicles (UAV).”
Zhou, H., Lynch, J. P., & Zekkos, D. (2022) · Structural Control and Health Monitoring · doi:10.1002/stc.2942 · Cited by 13
Large-Scale Dense Mapping System Based on Visual-Inertial Odometry and Densely Connected U-Net
Fan, C., Hou, J., & Yu, L. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3250301 · Cited by 10
“Forster et al.’s paper [51] describes in detail the handling of IMU biases, and Qin’s paper [12] describes the method of IMU preintegration.”
MethodA Survey of Simultaneous Localization and Mapping with an Envision in 6G Wireless Networks
Huang, B., Zhao, J., Luo, S., & Liu, J. (2021) · Journal of Global Positioning Systems · doi:10.5081/jgps.17.2.206 · Cited by 13
Invisible QR Code Hijacking Using Smart LED
Zhou, A., Su, G., Zhu, S., & Ma, H. (2019) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3351284 · Cited by 10
“, wireless radio [16, 55, 62, 64] or imaging [20, 21] based localization), VLP has the desirable advantages of centimeter-grained precision and high robustness to environment dynamics (e.”
BackgroundRapid and Robust Monocular Visual-Inertial Initialization with Gravity Estimation via Vertical Edges
Li, J., Bao, H., & Zhang, G. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8968456 · Cited by 14
“A reader may refer to [15] for details of the random-walk model of IMU biases, as well as their units.”
BackgroundMethodInfluentialVehicle velocity estimation based on WSS/IMU with wheel slip recognition
Yuan, L., Chen, H., Wang, Y., & Lian, X. (2023) · Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering · doi:10.1177/09544070231181163 · Cited by 9
Benchmark for Evaluating Initialization of Visual-Inertial Odometry
Zhao, Z., & Chen, B. M. (2023) · Chinese Control Conference (CCC) · doi:10.23919/ccc58697.2023.10240505 · Cited by 24
“Considering the IMU measurements between two time consecutive visual frames, i and i + 1 , the residual for preintegrated IMU measurement from [21] can be written as A standard model for the visual residual error of a single image measurement z ij is the reprojection error ) where π ( · ) is the…”
MethodGlobal principal planes aided LiDAR-based mobile mapping method in artificial environments
Bao, S., Shi, W., Yang, D., Xiang, H., & Yu, Y. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2024.102472 · Cited by 10
NINT: Neural Inertial Navigation Based on Time Interval Information in Underwater Environments
He, Q., Yu, H., & Fang, Y. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3401008 · Cited by 9
Bai, X., & Hsu, L. (2024) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2024.3412208 · Cited by 14
Benchmarking of Laser-Based Simultaneous Localization and Mapping Methods in Forest Environments
Wang, X., Liang, X., Campos, M., Zhang, J., & Wang, Y. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3439438 · Cited by 12
“The IMU odometry was implemented through on-manifold pre-integration [61].”
MethodWang, Y., Tian, Y., Chen, J., Chen, C., Xu, K., & Ding, X. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3509541 · Cited by 10
“The preintegration of IMU measurements [46] between two consecutive image frames i and j is performed to derive the transformation involving the following steps : where R is rotation, t is translation, v is linear velocity, ω is angular velocity, a is acceleration, ˜ ω and ˜ a are the measurements…”
Vial, P., Solà, J., Palomeras, N., & Carreras, M. (2024) · IEEE Transactions on Robotics · doi:10.1109/tro.2024.3521865 · Cited by 9
Lei, D., Luo, L., Chen, S., & Li, W. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3522670 · Cited by 13
Building Forest Inventories With Autonomous Legged Robots—System, Lessons, and Challenges Ahead
Mattamala, M., Chebrolu, N., Frey, J., Freißmuth, L., Oh, H., Casseau, B., Hutter, M., & Fallon, M. F. (2025) · IEEE transactions on field robotics. · doi:10.1109/tfr.2025.3583972 · Cited by 10
“The system performs sensor fusion of relative pose displacements from LiDAR using the iterative closest point (ICP) algorithm [58], aided by preintegrated IMU measurements [59].”
Air-Ground Collaboration for Language-Specified Missions in Unknown Environments
Cladera, F., Ravichandran, Z., Hughes, J., Murali, V., Nieto-Granda, C., Hsieh, M. A., Pappas, G. J., Taylor, C. J., & Kumar, V. (2025) · IEEE transactions on field robotics. · doi:10.1109/tfr.2025.3584019 · Cited by 28
“We get 5-Hz global position readings from the GNSS, which we model as unary pose factors [81] and use the preintegrated IMU measurements [92] to constrain the pose graph.”
Parameter Estimation of an Industrial Car-Like Tractor
Zhao, H., Chen, W., Zhou, S., & Liu, Y. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3068943 · Cited by 9
“With the involvement of vehicle dynamic constraints, the IMU preintegration technique [23] can no longer be applied and we need to add a new state at each IMU measurement.”
BackgroundMethodInfluentialGraph-optimisation-based self-calibration method for IMU/odometer using preintegration theory
Bai, S.-Y., Lai, J., Lyu, P., Cen, Y., Wang, B.-Q., & Sun, X. (2022) · Journal of navigation · doi:10.1017/s0373463321000722 · Cited by 10
“To solve this problem, preintegration theory (Forster et al., 2017; Yuan et al., 2018) was introduced into the odometer.”
MethodNovel robust simultaneous localization and mapping for long-term autonomous robots
Wei, W., Zhu, X., & Wang, Y. (2022) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.2000358 · Cited by 10
“We generate relative IMU measurements be-tween S i and S j through the IMU pre-integration method (Forster et al., 2017).”
MethodA SLAM System with Direct Velocity Estimation for Mechanical and Solid-State LiDARs
Jie, L., Jin, Z. J., Wang, J., Zhang, L., & Tan, X. (2022) · Remote Sensing · doi:10.3390/rs14071741 · Cited by 9
“The continuous factor is similar to the IMU pre-integration factor in VIO [39] and imposes a constraint on the motion of adjacent scans.”
BackgroundDing, J., Huang, C., Cheng, J., Wang, F., & Hu, Y. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3233966 · Cited by 10
“In the “two-sample” algorithm, ˜ ω k and ˜ a k are computed as follows : while, in the work of [10] and [24], ˜ ω k and ˜ a k are equal to computed as follows : Finally, the refined preintegrated IMU measurements can be computed as follows : where ˜ ω k and ˜ a k in (35) are computed by (30) or…”
BackgroundMethodInfluentialYan, D., Shi, C., Li, T., Jiang, H., Xia, M. Y., Guo, Q., & Li, Y. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3351757 · Cited by 10
“The IMU pre-integration method [48] is employed to derive the short-term increments of the body motion state.”
MethodA Comparative Analysis of Visual-Inertial SLAM for Assisted Wayfinding of the Visually Impaired
Zhang, H., Jin, L., Zhang, H., & Ye, C. (2019) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2019.00028 · Cited by 8
“According to [20], the SO(3) manifold based IMU pre-integration method outperform Euler angle based one.”
MethodFast and Robust Monocular Visua-Inertial Odometry Using Points and Lines
Zhang, N., & Zhao, Y. (2019) · Sensors · doi:10.3390/s19204545 · Cited by 10
“in 2016 [40], which uses the IMU pre-integration method to transform IMU measurement data into visual keyframe constraints.”
MethodThe First Attempt of SAR Visual-Inertial Odometry
Liu, J., Qiu, X., & Ding, C. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.2995774 · Cited by 10
Direct Sparse Stereo Visual-Inertial Global Odometry
Wang, Z., Li, M., Zhou, D., & Zheng, Z. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561410 · Cited by 10
“When a new keyframe is created, all accelerometer and gyroscope data between two consecutive keyframes are pre-integrated following to [25].”
BackgroundMethodZhang, J., Wen, W., Huang, F., Chen, X., & Hsu, L. (2021) · Proceedings of the Satellite Division's International Technical Meeting (Online)/Proceedings of the Satellite Division's International Technical Meeting (CD-ROM) · doi:10.33012/2021.18089 · Cited by 9
Sharing visual-inertial data for collaborative decentralized simultaneous localization and mapping
Dubois, R., Eudes, A., & Frémont, V. (2021) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2021.103933 · Cited by 16
Jeong, J., Mishra, H., Ott, C., & Kim, M. J. (2022) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra46639.2022.9812268 · Cited by 9
A Fast Analytical Two-Stage Initial-Parameters Estimation Method for Monocular-Inertial Navigation
Wei, H., Zhang, T., & Zhang, L. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3210967 · Cited by 12
“Generally, the frequency of the IMU outputs is faster than the camera exposure, therefore, the preintegration model proposed by [10] and [11] can be used to synchronize IMU and camera frequency.”
BackgroundMethodInfluentialAn Energy Efficient and Runtime Reconfigurable Accelerator for Robotic Localization
Liu, Q., Hao, Y.-H., Liu, W.-Z., Yu, B., Gan, Y.-M., Tang, J., Liu, S.-S., & Zhu, Y. (2023) · IEEE transactions on computers · doi:10.1109/tc.2022.3230899 · Cited by 10
A Robust Lidar SLAM System Based on Multi-Sensor Fusion
Zhang, F., Zhang, B., & Sun, C. (2022) · International Conference on Control, Automation and Information Sciences (ICCAIS) · doi:10.1109/iccais56082.2022.9990085 · Cited by 11
“The computing formula and principle of the pre-integration measurements are introduced in detail in [16].”
MethodKim, B., Azhari, M. B., Park, J.-Y., & Shim, D. H. (2024) · Journal of Field Robotics · doi:10.1002/rob.22284 · Cited by 14
“…δ R t t − 1, , and surrounding information Ensure : The current UAV state T t , the optimized UAV state T * , and the relative transformation δ T t 1: t − 1, is the relative motion increment of rotation between time t − 1 and t , computed using the IMU preintegration method (Forster et al., 2017)”
MethodElisha, Y. B., & Indelman, V. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8206085 · Cited by 9
Xia, L., Meng, Q., Chi, D., Meng, B., & Yang, H. (2019) · Sensors · doi:10.3390/s19092004 · Cited by 9
“The ( i + 1 ) th updated R i + 1 WB , W v i + 1 B and W p i + 1 B can be given by [29]: where ∆ t i , i + 1 is the time interval between two keyframes.”
BackgroundA Review of Visual Odometry Methods and Its Applications for Autonomous DrivingPreprint
Lim, K. L., & Bräunl, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2009.09193 · Cited by 19
“The authors then analysed these challenges using a feature-based method, a semi-direct method, and a direct method, which are ORB-SLAM [25], SVO [26] and DSO [27] respectively.”
MethodInfluentialVI-NeRF-SLAM: a real-time visual–inertial SLAM with NeRF mapping
Liao, D., & Ai, W. (2024) · Journal of Real-Time Image Processing · doi:10.1007/s11554-023-01412-6 · Cited by 14
“Following the theory developed in [22]and formulated on manifolds in [12],the pre-integrated IMU increments can be denoted as ∆ R i,j , ∆ v i,j , ∆ p i,j .”
SlimSLAM: An Adaptive Runtime for Visual-Inertial Simultaneous Localization and Mapping
Behroozi, A., Chen, Y., Fruchter, V., Subramanian, L., Srikanth, S., & Mahlke, S. A. (2024) · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3620666.3651361 · Cited by 11
“Suf-fixes Suf-fixes "dt", "rpi", and "eCPU" denote the hardware platforms: desktop, Raspberry Pi 4, and embedded CPU respectively. algorithms [15, 55, 61, 62] out-of-the-box but significantly exceed the power target.”
BackgroundRobust Radar Inertial Odometry in Dynamic 3D Environments
Lyu, Y., Hua, L., Wu, J., Liang, X., & Zhao, C. (2024) · Drones · doi:10.3390/drones8050197 · Cited by 12
TC$^{2}$LI-SLAM: A Tightly-Coupled Camera-LiDAR-Inertial SLAM System
Tong, Y., Zhang, X., Wang, R., Song, Z., Wu, S., Zhang, S., Wang, Y., & Yuan, J. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3426375 · Cited by 8
SaDVIO: Sparsify and Densify VIO for UGV Traversability Estimation
Debeunne, C., Torres, A., & Vivet, D. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3436339 · Cited by 8
Robust High-Speed State Estimation for Off-Road Navigation Using Radar Velocity Factors
Nissov, M., Edlund, J. A., Spieler, P., Padgett, C. W., Alexis, K., & Khattak, S. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3486189 · Cited by 15
“This phasedifferencerelates to the doppler measurement v r (also called radial speed) of a given object as a function of the transmit time of a single chirp and the wavelength.”
BackgroundCo-RaL: Complementary Radar-Leg Odometry with 4-DoF Optimization and Rolling Contact
Jung, S., Yang, W., & Kim, A. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10801960 · Cited by 8
“To avoid the recomputation of ∆˜ p ij for every input, we exploit the first order approximation as b = ¯ b + δ b following [22].”
A Fast MEMS-IMU/GPS In-Motion Alignment Method Using Full-Integration-Based Position Loci
Zhou, X., Zhang, M., Hu, J., Wu, C., & Guan, X. (2025) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2025.3563700 · Cited by 8
“The inertial measurement unit (IMU), as a critical sensor of the SINS, typically consists of a three-axis gyroscope and accelerometer module [5], [6], [7], [8].”
Tsao, S.-H., & Jan, S. (2020) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3046718 · Cited by 8
“Among the literature in the field of computer vision and robotics, the manifold and the matrix Lie group representations for rigid body motion have been applied to handle the uncertainty in the case of both filtering-based algorithms [21]–[23] and optimization-based algorithms [24], [25].”
MethodLIDAR based distance estimation for emergency use terrestrial autonomous robot
Bălașa, R.-I., Olaru, G., Constantin, D., Stefan, A. G., Bilu, C.-M., & Balaceanu, M. B. (2021) · International Conference on Electronics, Computers and Artificial Intelligence (ECAI) · doi:10.1109/ecai52376.2021.9515047 · Cited by 8
Zhang, Q., Hao, L., Li, H., Ren, Z., & Cheng, J. S. (2022) · International Symposium on Autonomous Systems (ISAS) · doi:10.1109/isas55863.2022.9757267 · Cited by 10
DOES: A Deep Learning-Based Approach to Estimate Roll and Pitch at Sea
Ciaccio, F. D., Russo, P., & Troisi, S. (2022) · IEEE Access · doi:10.1109/access.2022.3158971 · Cited by 8
“Moreover, visual methods can be further integrated with traditional, IMU-based orientation estimation algorithms to obtain a robust and reliable visualinertial odometry system [7].”
MethodMa, T., Chen, S., Ruan, L., & Xu, Y. (2023) · International Conference on Automation, Control and Robotics Engineering (CACRE) · doi:10.1109/cacre58689.2023.10208719 · Cited by 8
“ij ∆p% are the predicted changes according to IMU measurements and estimated biases [12]: ( ) ( ) ( ) ( ) ( ) 1”
BackgroundMethodInfluentialLmapping: tightly-coupled LiDAR-inertial odometry and mapping for degraded environments
Zou, J., Shao, L., Tang, H., Chen, H., Bao, H., & Pan, X. (2023) · Intelligent Service Robotics · doi:10.1007/s11370-023-00482-6 · Cited by 8
“Forster et al. [18] proposed the well-known IMU pre-integration theory.”
BackgroundFusing Multiple Isolated Maps to Visual Inertial Odometry Online: A Consistent Filter
Zhang, Z., Jiao, Y., Xiong, R., & Wang, Y. (2023) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2023.3314080 · Cited by 8
“To reduce the number of variables in the optimization, IMU preintegration technique is proposed as a factor between the two keyframes [25].”
MethodSemantic-Assisted LIDAR Tightly Coupled SLAM for Dynamic Environments
Liu, P., Bi, Y., Shi, J., Zhang, T., & Wang, C. (2024) · IEEE Access · doi:10.1109/access.2024.3369183 · Cited by 8
“Subsequently, the IMU preintegration method proposed in [26] is employed to obtain the relative physical motion between two synchronized time instances.”
MethodVPE-SLAM: Virtual Point Enhanced SLAM Using Solid-State LiDAR for Weak Feature Environments
Chen, J., Zi, Y., Li, N., Guo, S., Hao, X., Cui, G., & Yang, X. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3382350 · Cited by 8
“S IMULTANEOUS localization and mapping (SLAM) are integral to a range of indoor robotic applications, encompassing robot surveying, control, and map reconstruction, as extensively discussed in various studies [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14].”
BackgroundJaquier, N., Rozo, L. D., & Asfour, T. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10611701 · Cited by 8
“Modeling complex, multimodal probability distributions on smooth manifolds is at the heart of various robotics tasks ranging from state estimation [12] to motion planning [13] and learning from demonstrations [6], [14].”
BackgroundIMU as an input versus a measurement of the state in inertial-aided state estimation
Burnett, K., Schoellig, A. P., & Barfoot, T. D. (2025) · Robotica · doi:10.1017/s0263574724002121 · Cited by 15
“We use classic preintegration to form binary factors between pairs of estimated states in the sliding window [2].”
InfluentialShop floor data-driven spatial–temporal verification for manual assembly planning
Fang, W., & Zheng, L. (2019) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-019-01491-y · Cited by 8
“2 The flowchart for the shop-floor data collection tial fusion pipeline (Forster et al. 2017).”
Autonomous Door Opening With the Interacting-BoomCopter Unmanned Aerial Vehicle
McArthur, D. R., Chowdhury, A. B., & Cappelleri, D. J. (2019) · Journal of Mechanisms and Robotics · doi:10.1115/1.4045843 · Cited by 8
ARAS: Ambiguity-aware Robust Active SLAM based on Multi-hypothesis State and Map Estimations
Hsiao, M., Mangelson, J. G., Suresh, S., Debrunner, C. H., & Kaess, M. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341384 · Cited by 12
“preintegration [7] is regarded as ambiguous if the estimated IMU bias is larger than a threshold.”
BackgroundZhao, M., Zhou, D., Song, X., Chen, X., & Zhang, L. (2022) · Sensors · doi:10.3390/s22093389 · Cited by 12
“In a visual-inertial SLAM system, IMU preintegration measurements [26] can be obtained by the measurement integration between consecutive image frames, i and i + 1, for rotation, velocity, and position with the formulation on manifolds [27], noted as ∆Ri,i+1, ∆vi,i+1, and ∆pi,i+1, respectively:”
Zhai, Y., & Zhang, S. (2022) · Sensors · doi:10.3390/s22134778 · Cited by 8
“The variance of the IMU pre-integration factor can be obtained iteratively from an initial value 0 using the method in [26].”
BackgroundMethodInfluentialFaizullin, M., & Ferrer, G. (2022) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2022.104316 · Cited by 9
A Lidar-Inertial Navigation System for UAVs in GNSS-Denied Environment with Spatial Grid Structures
Qiu, Z., Lv, J., Lin, D., Yu, Y., Sun, Z., & Zheng, Z. (2022) · Applied Sciences · doi:10.3390/app13010414 · Cited by 8
Equivariant Filter Design for Discrete-time Systems
Ge, Y., van Goor, P., & Mahony, R. E. (2022) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc51059.2022.9992342 · Cited by 8
Monocular visual-inertial odometry leveraging point-line features with structural constraints
Zhang, J., Yang, J., & Ma, J. (2023) · The Visual Computer · doi:10.1007/s00371-023-02807-z · Cited by 8
“The IMU preintegration method used in our system is proposed in [41].”
Robot Intelligence for Real World Applications
Liu, Y., Fan, Z., Guo, R., Wang, J., Nie, Q., Wang, X., & Wang, Z. (2018) · Chinese Journal of Electronics · doi:10.1049/cje.2018.03.007 · Cited by 11
Communication constrained cloud-based long-term visual localization in real time
Ding, X., Wang, Y., Tang, L., Yin, H., & Xiong, R. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8968550 · Cited by 8
“The preintegration error term e preint and the bias error term e bias for keyframe i and k is defined as where the error terms e R , e p , e v and e b are formulated as the rotational, positional, velocity and bias errors following [14].”
MethodTowards Globally Consistent Visual-Inertial Collaborative SLAM
Karrer, M., & Chli, M. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8461213 · Cited by 8
“For more detailed explanation of pre-integration the reader is kindly referred to [7].”
BackgroundMethodMonocular Visual-Inertial-Pressure fusion for Underwater localization and 3D mapping.
Ferrera, M. (2019) · theses.fr (ABES) · doi:10.70675/0a530b06ze9c8z4a9dzabccz9acc11efeb31 · Cited by 13
“…UW-VO à des méthodes de l’état de l’art en SLAM visuel monoculaire, à savoir ORB-SLAM (Mur-Artal et al., 2015), LSD-SLAM (Engel et al., 2014) et SVO (Forster et al., 2017), et nous avons montré qu’UWVO est plus robuste et plus précis que ces méthodes sur des séquences vidéos prises en milieux…”
MethodInfluentialFangbin, 汪. W., Kun, 曹. C., Xue, 龚. G., Darong, 朱. Z., & Ping, 汪. W. (2024) · Laser & Optoelectronics Progress · doi:10.3788/lop232263 · Cited by 7
Li, Y., Liu, F., Zhang, J., Wang, J., Han, H., & Hao, C. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3487980 · Cited by 8
Invariant-DLIO: Direct LiDAR–Inertial Odometry Based on Invariant Kalman Filtering
Fang, K., Song, R., & Ho, I. W. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3558916 · Cited by 8
“In this LC-LIO variant, the InEKF propagation and correction modules were omitted, with initial values for the Incremental Generalized ICP (GICP) algorithm instead derived from IMU preintegration [45].”
Enhancing Visual–Inertial Odometry Robustness and Accuracy in Challenging Environments
Minervini, A., Carrio, A., & Guglieri, G. (2025) · Robotics · doi:10.3390/robotics14060071 · Cited by 10
Okawara, T., Koide, K., Takanose, A., Oishi, S., Yokozuka, M., Uno, K., & Yoshida, K. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3580332 · Cited by 7
“• Unitree odometry w/ LIO : Matching cost factor [16], IMU pre-integration factor [17], and the above Unitree odometry algorithm-based motion constraint are incorporated into a factor graph for reasonable comparison.”
Adaptive Covariance and Quaternion-Focused Hybrid Error-State EKF/UKF for Visual-Inertial Odometry
Asıl, U., & Nasibov, E. N. (2025) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-025-00942-0 · Cited by 12
Du, T., Shan, D., & Guo, P. F. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3587482 · Cited by 7
“Due to space limitations, the detailed derivation of (17)–(26) are given in [31] 2) LiDAR Odometry Factor: The LiDAR odometry factor is designed based on the LIO-SAM model.”
Wang, N., Zhao, H., & Chen, T. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3601970 · Cited by 10
GaRLIO: Gravity Enhanced Radar-LiDAR-Inertial Odometry
Noh, C., Yang, W., Jung, M., Jung, S., & Kim, A. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11128334 · Cited by 7
“However, establishing the relationship between pose and acceleration generally results in amplified errors due to the double integrated bias and noise of the IMU acceleration [10].”
Kong, A., Wang, Y., & Zhao, J. (2025) · Measurement · doi:10.1016/j.measurement.2025.118983 · Cited by 10
Zhang, Z., Sheng, C., Pan, S., Wang, X., Yu, B., & Zhang, J. (2025) · Remote Sensing · doi:10.3390/rs17223760 · Cited by 7
“The initial frame of lidar odometry is initialized using the rotation and translation obtained from IMU pre-integration [27].”
GPS‐Denied LiDAR‐Based SLAM—A Survey
Jiang, H., Cheng, Y., Dai, W., Wan, W., Liu, Q., & Wang, F. (2025) · IET Cyber-Systems and Robotics · doi:10.1049/csy2.70031 · Cited by 17
An Emergency Positioning System Fusing GEO Satellite Doppler Observation and INS for SOTM
Yuan, D. Z., Yong-yuan, Q., Wu, Z., & Shen, X. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3096568 · Cited by 7
GNSS-Denied Semi-Direct Visual Navigation for Autonomous UAVs Aided by PI-Inspired Inertial Priors
Gallo, E., & Barrientos, A. (2023) · Aerospace · doi:10.3390/aerospace10030220 · Cited by 8
LVIF: a lightweight tightly coupled stereo-inertial SLAM with fisheye camera
Zhu, H., Zhang, G., Ye, Z., & Zhou, H. (2023) · Complex & Intelligent Systems · doi:10.1007/s40747-023-01190-5 · Cited by 10
“Note that for the completeness of the article, some equations presented are not contribution of this work, mainly from [38].”
Lee, D., Nam, H., Ryu, C., Nah, S., Moon, S., & Shim, D. H. (2024) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2024.3366153 · Cited by 9
“This predictive approach leverages the motion estimation from LiDAR data, providing a more informed starting point for the iterative scan matching process.”
MethodLeng, J., Sun, C., Wang, B., Lan, Y., Huang, Z., Zhou, Q., Liu, J., & Li, J. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3415170 · Cited by 10
“In addition to the feature-based method and direct method, SVO [37] is a semidirect method that combines both feature-based and direct methods.”
Method
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.