On the comparison of uncertainty criteria for active SLAM
Open accessThe DOI matches its Crossref record; 17 of the 18 references it cites were found in a registry or an index.
Abstract
In this paper, we consider the computation of the D-optimality criterion as a metric for the uncertainty of a SLAM system. Properties regarding the use of this uncertainty criterion in the active SLAM context are highlighted, and comparisons against the A-optimality criterion and entropy are presented. This paper shows that contrary to what has been previously reported, the D-optimality criterion is indeed capable of giving fruitful information as a metric for the uncertainty of a robot performing SLAM. Finally, through various experiments with simulated and real robots, we support our claims and show that the use of D-opt has desirable effects in various SLAM related tasks such as active mapping and exploration.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Bajcsy, R. (1988) · Proceedings of the IEEE · doi:10.1109/5.5968 · Cited by 830 · cited 1×
“As mentioned above, the integration of trajectory planning or, what is equivalent, applying the active sensing paradigm [16] [10] to the SLAM problem, involves the optimization of a multi-objective performance criterion or value function J .”
MethodGeneral Equivalence Theory for Optimum Designs (Approximate Theory)
Kiefer, J. (1974) · The Annals of Statistics · doi:10.1214/aos/1176342810 · Cited by 761 · cited 9×
“This solution rises naturally if the D-opt is derived from the family of uncertainty criteria proposed by Kiefer in [15], This family of uncertainty criteria is valid in the range of 0 p ∞ for a covariance matrix ( Σ ) of size l × l associated to a design ξ ( e.g. π ).”
BackgroundMethodInfluentialPukelsheim, F. (2006) · Society for Industrial and Applied Mathematics eBooks · doi:10.1137/1.9780898719109 · Cited by 684
Information Gain-based Exploration Using Rao-Blackwellized Particle Filters
Stachniss, C., Grisetti, G., & Burgard, W. (2005) · Robotics: Science and Systems I · doi:10.15607/rss.2005.i.009 · Cited by 542 · cited 1×
“[2] and [3] proposed a discrete and greedy planning methodology.”
BackgroundAdaptive Mobile Robot Navigation and Mapping
Feder, H. J. S., Leonard, J. J., & Smith, C. M. (1999) · The International Journal of Robotics Research · doi:10.1177/02783649922066484 · Cited by 362 · cited 2×
“The integration of the trajectory planning task into SLAM was first proposed in [1] and the term active SLAM referring to the aforementioned integration was coined by [8].”
BackgroundMethodAn experiment in integrated exploration
Makarenko, A. A., Williams, S. B., Bourgault, F., & Durrant-Whyte, H. F. (2003) · IEEE/RSJ International Conference on Intelligent Robots and System · doi:10.1109/irds.2002.1041445 · Cited by 304 · cited 1×
“[2] and [3] proposed a discrete and greedy planning methodology.”
BackgroundGlobal A-Optimal Robot Exploration in SLAM
Sim, R., & Roy, N. (2006) · Proceedings of the 2005 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.2005.1570193 · Cited by 238 · cited 5×
“2d shows an example of computing the determinant of the covariance matrix as reported in [10], [11] or [12], as can be seen after few steps -in this case 8- the value of the criterion goes to zero.”
BackgroundMethodInfluentialMartinez-Cantin, R., de Freitas, N., Brochu, E., Castellanos, J., & Doucet, A. (2009) · Autonomous Robots · doi:10.1007/s10514-009-9130-2 · Cited by 217 · cited 2×
“Recently, a continuous planning approach in states and actions has been proposed by [6].”
BackgroundSelf-citationTrajectory Optimization using Reinforcement Learning for Map Exploration
Kollar, T., & Roy, N. (2008) · The International Journal of Robotics Research · doi:10.1177/0278364907087426 · Cited by 139 · cited 2×
“Continuous states planning but with a discretization in actions space is explored in [5].”
BackgroundActive SLAM using Model Predictive Control and Attractor based Exploration
Leung, C., Huang, S., & Dissanayake, G. (2006) · 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2006.282530 · Cited by 112 · cited 2×
“The integration of the trajectory planning task into SLAM was first proposed in [1] and the term active SLAM referring to the aforementioned integration was coined by [8].”
BackgroundPukelsheim, F. (2006) · Society for Industrial and Applied Mathematics eBooks · Cited by 93
Multi-Step Look-Ahead Trajectory Planning in SLAM: Possibility and Necessity
Huang, S., Kwok, N. M., Dissanayake, G., Ha, Q. P., & Fang, G. (2006) · Proceedings of the 2005 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.2005.1570261 · Cited by 81 · cited 3×
“It is also possible to plan several steps ahead that will result, as has been pointed out by [4], in a faster convergence of the active SLAM goals but with an increase in the complexity of the computation.”
BackgroundInfluentialA Comparison of Decision Making Criteria and Optimization Methods for Active Robotic Sensing
Mihaylova, L. S., Lefebvre, T., Bruyninckx, H., Gadeyne, K., & Schutter, J. D. (2003) · Lecture notes in computer science · doi:10.1007/3-540-36487-0_35 · Cited by 70 · cited 6×
“2d shows an example of computing the determinant of the covariance matrix as reported in [10], [11] or [12], as can be seen after few steps -in this case 8- the value of the criterion goes to zero.”
BackgroundMethodInfluentialAction Selection for Single-Camera SLAM
Vidal‐Calleja, T., Sanfeliu, A., & Andrade‐Cetto, J. (2010) · IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · doi:10.1109/tsmcb.2010.2043528 · Cited by 38 · cited 1×
“Recently, Vidal-Calleja et al [17] intuited this, and proposed a solution that needs to suppose the maximum number of landmarks in the environment and initialize its covariance with a constant number.”
BackgroundTask Planning With Active Sensing For Autonomous Compliant Motion
Lefebvre, T., Bruyninckx, H., & Schutter, J. D. (2005) · The International Journal of Robotics Research · doi:10.1177/0278364904050079 · Cited by 27 · cited 5×
“2d shows an example of computing the determinant of the covariance matrix as reported in [10], [11] or [12], as can be seen after few steps -in this case 8- the value of the criterion goes to zero.”
BackgroundMethodInfluentialUsing reinforcement learning to improve exploration trajectories for error minimization
Kollar, T., & Roy, N. (2006) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2006.1642211 · Cited by 25 · cited 1×
“The choice of this criterion in many active SLAM related works such as [7], [8], [5], [9], and [6], among others, had its foundation in the fact that papers such as [10], [11], and [12] reported that (i) the A-opt applied to the problems of planning under uncertainty out performs other well-known criteria such as the…”
BackgroundHeuristic search planning to reduce exploration uncertainty
Meger, D., Rekleitis, I. M., & Dudek, G. (2008) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2008.4651015 · Cited by 17 · cited 1×
“The choice of this criterion in many active SLAM related works such as [7], [8], [5], [9], and [6], among others, had its foundation in the fact that papers such as [10], [11], and [12] reported that (i) the A-opt applied to the problems of planning under uncertainty out performs other well-known criteria such as the…”
Background
Research works printed but not linked: 3
- olson, 0, Evaluating the performance of map optimization algorithms, Robotics Science and Systems Workshop on Good Experimental Methodology in Robotics June 2009
- kurlbaum, 2008, Dlr spatial cognition data set
- pa?zman, 1986, Foundations of Optimum Experimental Design (Mathematics and Its Applications)
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
“One of these frameworks is the Theory of Optimal Experimental Design (TOED) [198] which, applied to active SLAM [42], [44], allows selecting future robot action based on the predicted map uncertainty.”
Self-citationPast, 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
Self-citationA Survey on Active Simultaneous Localization and Mapping: State of the Art and New Frontiers
Placed, J. A., Strader, J., Carrillo, H., Atanasov, N., Indelman, V., Carlone, L., & Castellanos, J. A. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2023.3248510 · Cited by 332
“[24] showed these problems can be solved using Kiefer’s formulation ofD-optimality [151], thus re-establishing the latter as an effective measure of uncertainty for active SLAM.”
BackgroundMethodSelf-citationUncertainty-aware receding horizon exploration and mapping using aerial robots
Papachristos, C., Khattak, S., & Alexis, K. (2017) · 2017 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2017.7989531 · Cited by 185
“Following the unifying uncertainty criteria definitions of Kiefer [30], the particular formulation of [7] is employed: As discussed in [7], this formulation addresses the concerns regarding the capacity of D-opt to enable task completion checking, and its rate of being driven to zero.”
BackgroundMethodLearning Exploration Policies for Navigation
Chen, T., Gupta, S., & Gupta, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.01959 · Cited by 34
“Past works have formulated active SLAM as Partially Observable Markov Decision Processes (POMDPs) (Martinez-Cantin et al., 2009), or as choosing actions that reduce uncertainty in the estimated map (Carrillo et al., 2012).”
Background
Metadata
- Authors
- Carrillo, H., Reid, I., & Castellanos, J. A.
- Affiliations
- Universidad de Zaragoza (ES)
- University of Oxford (GB)
- Venue
- 2012 IEEE International Conference on Robotics and Automation
- Publisher
- IEEE
- Published
- 2012-05
- Pages
- 2080-2087
- ISSN
- 1050-4729 (ISSN-L)
- OpenAlex
- W2161123599
- Cited by
- 144
- References
- 18
- Citations by year
- 2026: 2 · 2025: 11 · 2024: 13 · 2023: 13 · 2022: 15 · 2021: 16 · 2020: 12 · 2019: 12 · 2018: 12 · 2017: 6 · 2016: 9 · 2015: 8 · 2014: 7 · 2013: 7 · 2012: 1
- FWCI
- 13.71
- Type
- conference paper
- Language
- en
- Open access
- green
- zaguan.unizar.es
- Licence
- cc-by-nc-sa
- Locations
- Proceedings - IEEE International Conference on Robotics and Automation/Proceedings (publishedVersion)
- Adelaide Research & Scholarship (AR&S) (University of Adelaide) (submittedVersion)
- LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) (submittedVersion, cc-by-nc-sa)
- Zaguan (Universidad de Zaragoza) (submittedVersion, cc-by-nc-sa)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Artificial Intelligence › Target Tracking and Data Fusion in Sensor Networks
- Keywords
- active SLAM, uncertainty measures, entropy (information theory), robot exploration
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.