A hierarchical wavelet decomposition for continuous-time SLAM
The DOI matches its Crossref record; 19 of the 20 references it cites were found in a registry or an index.
Abstract
This paper proposes using hierarchical wavelets as a basis in parametric continuous-time batch estimation. The need for a continuous-time robot pose in the simultaneous localization and mapping (SLAM) problem has arisen as state-of-the-art batch SLAM algorithms attempt to handle more challenging hardware; specifically, the continuous-time framework is particularly beneficial when using high-rate sensors, multiple unsynchronized sensors, or scanning sensors, such as lidar and rolling-shutter cameras, during motion. Although the traditional discrete-time SLAM formulation can be adapted by using temporal pose interpolation, approaches using the continuous-time framework are able to generate smooth robot trajectories with less state variables. In this paper, we focus on the parametric approach using temporal basis functions to develop a finite-element representation of the continuous-time robot trajectory. While the majority of current implementations have utilized a uniformly spaced B-spline basis, we note that trajectory richness is often quite variable; in this paper, we show how a hierarchical system of wavelet basis functions can be used to increase the resolution of the solution only in the temporally local regions of the trajectory that require additional detail. We validate our approach by contrasting uniform B-splines and wavelets in a six-dimensional pose-graph SLAM experiment, using both simulated and real data.
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 cited source says what the claim attributes to it.
In the citing paper
“Later, Anderson et al. [4] and Dub´ e et al. [67] proposed more efficient implementations by using wavelets or sampling non-uniform knots over the trajectory, respectively.”
From Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016) · cites this paper as [4]
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
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×
“In particular, the use of stereo cameras for visual odometry remains a leading paradigm that encourages the formulation of a full nonlinear least-squares optimization problem, because it is both flexible and efficient [1].”
Biorthogonal bases of compactly supported wavelets
Cohen, A., Daubechies, I., & Feauveau, J. ‐. (1992) · Communications on Pure and Applied Mathematics · doi:10.1002/cpa.3160450502 · Cited by 2,667 · cited 1×
“In this extreme case, the entries corresponding to scaling functions and wavelets in the levels (cid:96) = 0 ... 2 are fully dense because they have local support over the entire trajectory; using more time segments at the coarsest level will result in a sparser upper and left hand side.”
BackgroundWavelets for computer graphics: theory and applications
Stollnitz, E. J., DeRose, T., & Salesin, D. (1996) · Morgan Kaufmann Publishers Inc. eBooks · Cited by 696
Zebedee: Design of a Spring-Mounted 3-D Range Sensor with Application to Mobile Mapping
Bosse, M., Zlot, R., & Flick, P. (2012) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2200990 · Cited by 399 · cited 1×
“The necessary background information required to develop a hierarchical wavelet basis for continuous-time state estimation is summarized here.”
BackgroundAssociating Uncertainty With Three-Dimensional Poses for Use in Estimation Problems
Barfoot, T. D., & Furgale, P. T. (2014) · IEEE Transactions on Robotics · doi:10.1109/tro.2014.2298059 · Cited by 305
Self-citationContinuous 3D scan-matching with a spinning 2D laser
Bosse, M., & Zlot, R. (2009) · 2009 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.2009.5152851 · Cited by 289 · cited 1×
“Using a continuously spinning lidar, Bosse and Zlot [5] present a technique for performing SLAM while scanning and moving; the technique is akin to the iterative closest point (ICP) algorithm and interpolates poses in order to compensate a sequence of temporally discretized lidar scans.”
MethodContinuous-time batch estimation using temporal basis functions
Furgale, P., Barfoot, T. D., & Sibley, G. (2012) · 2012 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2012.6225005 · Cited by 259 · cited 6×
“Furgale et al. [2] formally derive the general continuous-time SLAM problem and demonstrate the use of uniformly spaced temporal basis functions to model the state in a typical camera-IMU calibration problem.”
BackgroundMethodResultInfluentialSelf-citationRolling Shutter Camera Calibration
Oth, L., Furgale, P., Kneip, L., & Siegwart, R. (2013) · 2013 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2013.179 · Cited by 150 · cited 2×
“Recent work by Oth et al. [14] demonstrates the use of parametric, continuous-time batch estimation to calibrate the line delay of a rolling-shutter camera.”
MethodAdaptive relative bundle adjustment
Sibley, G., Mei, C., Reid, I., & Newman, P. (2009) · Robotics: Science and Systems V · doi:10.15607/rss.2009.v.023 · Cited by 139 · cited 1×
“Anderson and Barfoot [12] also leverage this parametric approach to continuous-time estimation, but derive the relative formulation of the problem, similar to the discrete-time work of Sibley et al. [13].”
ResultRolling shutter bundle adjustment
Hedborg, J., Forssen, P.-E., Felsberg, M., & Ringaby, E. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6247831 · Cited by 129 · cited 1×
“With a high-rate, rolling-shutter camera, Hedborg et al. [9] demonstrate the application of interpolation to the full nonlinear bundle adjustment problem.”
Background
Research works printed but not linked: 1
- [19] S. Anderson, C. McManus, H. Dong, E. Beerepoot, and T. D. Barfoot, 'The gravel pit lidar-intensity imagery dataset,' Technical Report ASRL-2012-ABL001, UTIAS, 2012. -
Cited by
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I., & Leonard, J. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2016.2624754 · Cited by 3,596
“If SLAM has to work “out of the box” in arbitrary scenarios, methods for automatic tuning of the involved parameters need to be considered.”
Past, present, and future of simultaneous localization and mapping: Toward the robust-perception age
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I. D., & Leonard, J. J. (2016) · Zurich Open Repository and Archive (University of Zurich) · doi:10.5167/uzh-138892 · Cited by 3,048
Continuous-time Gaussian process motion planning via probabilistic inference
Mukadam, M., Dong, J., Yan, X., Dellaert, F., & Boots, B. (2018) · The International Journal of Robotics Research · doi:10.1177/0278364918790369 · Cited by 224
Self-citationContinuous-Time Visual-Inertial Odometry for Event Cameras
Mueggler, E., Gallego, G., Rebecq, H., & Scaramuzza, D. (2018) · IEEE Transactions on Robotics · doi:10.1109/tro.2018.2858287 · Cited by 193
“Cubic splines [7] or, in more general, Wavelets [8] are common basis functions for continuous-time”
BackgroundAnderson, S., & Barfoot, T. D. (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353368 · Cited by 132
“The parametric approaches use a finite number of temporal basis functions [21], [22], [23], [24], [25], [7] with the trajectory parametrized by the associated basis-function weights.”
MethodSelf-citationContinuous-time batch trajectory estimation using temporal basis functions
Furgale, P., Tong, C. H., Barfoot, T. D., & Sibley, G. (2015) · The International Journal of Robotics Research · doi:10.1177/0278364915585860 · Cited by 79
“First, the proposed methods for adaptively choosing the number of basis functions (Anderson et al., 2014; Oth et al., 2013) were both considered in the offline batch context.”
MethodSelf-citationContinuous-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
“Other suitable choices of basis functions are wavelets [13].”
BackgroundContinuous-Time Trajectory Estimation for Event-based Vision Sensors
Mueggler, E., Gallego, G., & Scaramuzza, D. (2015) · Robotics: Science and Systems XI · doi:10.15607/rss.2015.xi.036 · Cited by 49
“Wavelets have also been considered as basis for continuoustime trajectories [3].”
BackgroundCoco-LIC: Continuous-Time Tightly-Coupled LiDAR-Inertial-Camera Odometry Using Non-Uniform B-Spline
Lang, X., Chen, C., Tang, K., Ma, Y., Lv, J., Liu, Y., & Zuo, X. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3315542 · Cited by 56
“In real practice, the smoothness of a trajectory can vary significantly, and each segment actually requires a different density of control points to accurately model the real trajectory [34].”
BackgroundPöppl, F., Neuner, H., Mandlburger, G., & Pfeifer, N. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2022.12.022 · Cited by 43
Continuous-Time State Estimation Methods in Robotics: A Survey
Talbot, W., Nubert, J., Tuna, T., Cadena, C., Dümbgen, F., Tordesillas, J., Barfoot, T. D., & Hutter, M. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3593079 · Cited by 59
“Anderson et al. [50] used this idea to decide when to increase the active resolution level in their hierarchical formulation, based on work by Gortler and Cohen [134] who applied wavelet detail functions over a base B-spline signal function.”
Self-citationTrajectory representation and landmark projection for continuous-time structure from motion
Ovrén, H., & Forssén, P.-E. (2019) · The International Journal of Robotics Research · doi:10.1177/0278364919839765 · Cited by 38
“A popular choice of trajectory representation that is both C(2)-continuous and has compact support is cubic B-splines (Anderson et al. 2014; PatronPerez et al. 2015; Furgale et al. 2012; Lovegrove et al. 2013).”
BackgroundInfluentialDong, J., Mukadam, M., Boots, B., & Dellaert, F. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8461077 · Cited by 45
“Several popular continuous-time trajectory representations include linear interpolation [3], [4], [5], splines [6], [7], [8], [9], [10], and hierarchical wavelets [11].”
BackgroundSelf-citationYang, A., Cui, C., Bârsan, I. A., Urtasun, R., & Wang, S. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561481 · Cited by 37
“, piecewise linear functions [50], [56], spirals [57], wavelets [58], or B-splines [45], [59], [60].”
Localization of a mobile laser scanner via dimensional reduction
Lehtola, V., Virtanen, J., Vaaja, M. T., Hyyppä, H., & Nüchter, A. (2016) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2016.09.004 · Cited by 26
“Approaches include EKF-SLAM (Dissanayake et al., 2000), FastSLAM (Montemerlo et al., 2002), FastSLAM 2.0 (Montemerlo and Thrun, 2007), and GraphSLAM (Thrun and Montemerlo, 2006), including early approaches to 3D mapping (Thrun et al., 2000). Of these, GraphSLAM uses sparse matrices to represent a graph of observation…”
BackgroundMethodInfluentialSpline Error Weighting for Robust Visual-Inertial Fusion
Ovrén, H., & Forssén, P.-E. (2018) · 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/cvpr.2018.00041 · Cited by 22
MINS: Efficient and Robust Multisensor‐Aided Inertial Navigation System
Lee, W., Geneva, P., Chen, C., & Huang, G. P. (2025) · Journal of Field Robotics · doi:10.1002/rob.22546 · Cited by 36
“…uniform B ‐ spline (Furgale et al. 2013; Zheng et al. 2015; Droeschel and Behnke 2018; Cioffi et al. 2022; Lv et al. 2022), but there are other approaches, such as the Gaussian process (Barfoot 2017; Anderson et al. 2015), wavelets (Anderson et al. 2014), or Taylor series (Li and Mourikis 2014b).”
Anderson, S., MacTavish, K., & Barfoot, T. D. (2015) · The International Journal of Robotics Research · doi:10.1177/0278364915589642 · Cited by 19
“Temporal basis functions for continuous-time estimation are also used in the works of Anderson et al. (2014) and Oth et al. (2013), which investigate how parameterizations can be chosen adaptively to prevent underfitting and overfitting the measurements. Lovegrove et al. (2013) formulate a methodology for visual-inert…”
MethodInfluentialSelf-citationSparse Gaussian Processes for Continuous-Time Trajectory Estimation on Matrix Lie GroupsPreprint
Dong, J., Boots, B., & Dellaert, F. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.06020 · Cited by 23
“Several popular continuous-time trajectory representations include linear interpolation [10], [11], [12], splines [13], [14], [15], [16], [17], [18], and hierarchical wavelets [19], all of which have been used in both filtering and batch estimation approaches.”
BackgroundSelf-citationContinuous-Time Stereo-Inertial Odometry
Hug, D., Banninger, P., Alzugaray, I., & Chli, M. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3173705 · Cited by 21
Li, S., Li, X., Chen, S., Zhou, Y., & Wang, S. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3450654 · Cited by 18
Non-uniform sampling strategies for continuous correction based trajectory estimation
Dubé, R., Sommer, H., Gawel, A., Bosse, M. C., & Siegwart, R. (2016) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2016.7487683 · Cited by 15
“…τ a τ b , the componentwise exponentially linear interpolated Euclidean transformation T ( τ ) at τ ∈ [ τ a , τ b ] shall be the one uniquely given by where tra and rot denote the component projections as determined by the translate-after-rotate decomposition, i.e. ∀ T ∈ SE (3) T =: tra( T ) ◦…”
BackgroundContinuous latent state preintegration for inertial-aided systems
Gentil, C. L., & Vidal‐Calleja, T. (2023) · The International Journal of Robotics Research · doi:10.1177/02783649231199537 · Cited by 14
“More complex approaches use combinations of temporal basis functions such as B-splines or wavelets to provide the state with more smoothness and represen-tativeness (Anderson et al., 2014; Furgale et al., 2012).”
MethodContinuous-Time Visual-Inertial Trajectory Estimation with Event Cameras.Preprint
Mueggler, E., Gallego, G., Rebecq, H., & Scaramuzza, D. (2017) · arXiv (Cornell University) · Cited by 14
Continuous-time estimation for dynamic obstacle tracking
Ushani, A. K., Carlevaris‐Bianco, N., Cunningham, A. G., Galceran, E., & Eustice, R. M. (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353513 · Cited by 14
“Nonetheless, we will show that we can solve the obstacle tracking problem using a formulation similar to that of state of the art SLAM formulations.”
MethodLi, S., Chen, S., Li, X., Zhou, Y., & Wang, S. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103071 · Cited by 17
Flow-based SLAM: From geometry computation to learning
Yan, Z., & Zha, H. (2019) · Virtual Reality & Intelligent Hardware · doi:10.1016/j.vrih.2019.09.001 · Cited by 12
“Additional issues are also addressed in terms of noise model [131] and temporal density [138] .”
BackgroundA Spanning Tree-Based Multi-Resolution Approach for Pose-Graph Optimization
Tazaki, Y. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3185779 · Cited by 9
“Anderson et al. proposed a multi-resolution method based on the representation of posegraphs by wavelet transform [20].”
BackgroundMethodInfluentialHug, D., & Chli, M. (2020) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv50981.2020.00108 · Cited by 12
“The preconditions to allow efficient loop closing for CTSLAM problems in large scale environments were investigated in [12], where it was verified that moving to a relative representation is advantageous.”
BackgroundLi, X., Chen, S., Li, S., Zhou, Y., & Wang, S. (2023) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2023.3323278 · Cited by 7
“InadditiontoB-spline,theGaus-sianprocess(GP)regression[12]andhierarchicalwavelets[27] havealsoproventobefeasibleforcontinuous-timetrajectory modeling.”
BackgroundChen, S., Li, X., Li, S., Zhou, Y., & Wang, S. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3440092 · Cited by 5
Hyperion – A Fast, Versatile Symbolic Gaussian Belief Propagation Framework for Continuous-Time SLAM
Hug, D., Alzugaray, I., & Chli, M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73404-5_13 · Cited by 14
Practical Pose Trajectory Splines With Explicit Regularization
Persson, M., Hager, G., Ovrén, H., & Forssén, P.-E. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00026 · Cited by 7
B-Splined Trajectory Modified Generation to Maximize Speed of the Nonholonomic AMR Robot
Neamah, H. A., & Butdee, R. (2024) · Journal of Machine Engineering · doi:10.36897/jme/196037 · Cited by 4
“SLAM was used for temporal pose interpolation using continuous time framework for creating trajectory smoothly and using the parametric method of finite element representation [31].”
Rotation and Temporal Calibration of Event Camera and IMU Based on Contrast Maximization
Li, S., Li, X., Chen, S., & Zhou, Y. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3568945 · Cited by 3
“Comparatively, continuous-time implementations (e.g., B-spline [13], Gaussian process [14], hierarchical wavelets [15]) generally model the trajectory as the function of time, which holds the promise of seamlessly fusing multimodal sensors in a truly generic manner.”
eKalibr-Stereo: Continuous-Time Spatiotemporal Calibration for Event-Based Stereo Visual Systems
Chen, S., Li, X., & Yuan, L. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3608639 · Cited by 3
“This section presents the proposed event-based continuous-time stereo visual spatiotemporal calibration framework.”
Chen, S., Li, X., Li, S., Zhou, Y., & Wang, S. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3522622 · Cited by 2
“Compared to other continuous-time functions, e.g., Gaussian process [26], hierarchical wavelets [27], and Chebyshev interpolation [28], the uniform B-splines have closed-from analytic derivatives and local controllability, which could yield a sparse system and reduce computational complexity in…”
BackgroundCurve-graph odometry: Orientation-free error parameterisations for loop closure problems
Gutiérrez-Gómez, D., & Guerrero, J. J. (2015) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2015.07.017 · Cited by 2
VICO: Visual-Inertial Continuous-Time Odometry Based on Generalized Hermite Spline
Qi, H.-Y., Li, Z., Liu, H., Liu, X., & Deng, F. Y. (2025) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc65474.2025.11090719 · Cited by 1
“Overall, the continuous-time parameterization approaches can be classified into interpolation [4], [5], spline [6]–[9], hierarchical wavelets [10], and Gaussian processes [11], [12].”
Contact2Motion: Contact guided dexterous grasp motion generation with synergy embedded optimization
Ye, Q., Li, H., Liu, Q., Jiang, S., Zhou, T., Huo, Y., & Chen, J. (2025) · The International Journal of Robotics Research · doi:10.1177/02783649251364392 · Cited by 1
Visual Odometryin Principle and Practice
Persson, M. (2022) · Linköping studies in science and technology. Dissertations · doi:10.3384/9789179291693 · Cited by 1
Continuous Models for Cameras and Inertial Sensors
Ovrén, H. (2018) · Linköping studies in science and technology. Dissertations · doi:10.3384/diss.diva-148766 · Cited by 1
Soliman, A. (2023) · theses.fr (ABES) · doi:10.70675/32a52416z7918z4954z8b84ze91dda8e76dc
Metadata
- Authors
- Anderson, S., Dellaert, F., & Barfoot, T. D.
- Affiliations
- University of Toronto (CA)
- Georgia Institute of Technology (US)
- Venue
- 2014 IEEE International Conference on Robotics and Automation (ICRA)
- Publisher
- IEEE
- Published
- 2014-05
- Pages
- 373-380
- ISSN
- 1050-4729 (ISSN-L)
- OpenAlex
- W2065353791
- Cited by
- 44
- References
- 20
- Citations by year
- 2025: 7 · 2024: 5 · 2023: 5 · 2022: 4 · 2021: 2 · 2020: 1 · 2019: 3 · 2018: 5 · 2017: 2 · 2016: 4 · 2015: 6
- FWCI
- 6.73
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Proceedings - IEEE International Conference on Robotics and Automation/Proceedings (publishedVersion)
- SMARTech Repository (Georgia Institute of Technology) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Artificial Intelligence › Target Tracking and Data Fusion in Sensor Networks
- Physical Sciences › Engineering › Aerospace Engineering › Inertial Sensor and Navigation
- Keywords
- robot trajectory estimation, temporal basis functions, B-splines, pose graph SLAM
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.