A convex relaxation for approximate global optimization in simultaneous localization and mapping
Open accessThe DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Modern approaches to simultaneous localization and mapping (SLAM) formulate the inference problem as a high-dimensional but sparse nonconvex M-estimation, and then apply general first- or second-order smooth optimization methods to recover a local minimizer of the objective function. The performance of any such approach depends crucially upon initializing the optimization algorithm near a good solution for the inference problem, a condition that is often difficult or impossible to guarantee in practice. To address this limitation, in this paper we present a formulation of the SLAM M-estimation with the property that, by expanding the feasible set of the estimation program, we obtain a convex relaxation whose solution approximates the globally optimal solution of the SLAM inference problem and can be recovered using a smooth optimization method initialized at any feasible point. Our formulation thus provides a means to obtain a high-quality solution to the SLAM problem without requiring high-quality initialization.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Boyd, S., & Vandenberghe, L. (2004) · doi:10.1017/cbo9780511804441 · Cited by 56,601
(2006) · Springer Series in Operations Research and Financial Engineering · doi:10.1007/978-0-387-40065-5 · Cited by 9,008
Thrun, S. (2002) · Communications of the ACM · doi:10.1145/504729.504754 · Cited by 7,994 · cited 2×
“In this section we provide a brief review of the pose-and-landmark SLAM inference problem; interested readers are encouraged to consult [1], [9] for a more detailed presentation.”
BackgroundGolub, G. H., & Loan, C. F. V. (1996) · Johns Hopkins University Press eBooks · Cited by 6,027
Vandenberghe, L., & Boyd, S. (1996) · SIAM Review · doi:10.1137/1038003 · Cited by 4,105 · cited 1×
“it can be expressed as the solution set for a linear matrix inequality [10]).”
BackgroundLeast-squares estimation of transformation parameters between two point patterns
Umeyama, S. (1991) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.88573 · Cited by 2,420 · cited 1×
“…procedures of the form (29) have been studied previously, and there exist straightforward and efficient meth-ods to compute the mapping π R ( · ) (based upon the singular value decomposition [15] of ˜ R ) for the special case in which the norm (cid:107)·(cid:107) in (29) is the Frobenius norm [16].”
BackgroundG2o: A general framework for graph optimization
Kummerle, R., Grisetti, G., Strasdat, H., Konolige, K., & Burgard, W. (2011) · 2011 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2011.5979949 · Cited by 1,982 · cited 2×
“[13], [14], among others) will scale gracefully to problem sizes typical of those addressed by stateof-the-art least-squares methods for SLAM [3]–[5].”
BackgroundMethodiSAM2: Incremental smoothing and mapping using the Bayes tree
Kaess, M., Johannsson, H., Roberts, R., Ila, V., Leonard, J. J., & Dellaert, F. (2012) · The International Journal of Robotics Research · doi:10.1177/0278364911430419 · Cited by 1,362
Self-citationGlobally Consistent Range Scan Alignment for Environment Mapping
Lu, F., & Milios, E. (1997) · Autonomous Robots · doi:10.1023/a:1008854305733 · Cited by 1,304
An interior algorithm for nonlinear optimization that combines line search and trust region steps
Waltz, R. A., Morales, J. L., Nocedal, J., & Orban, D. (2006) · Mathematical Programming · doi:10.1007/s10107-004-0560-5 · Cited by 1,057 · cited 2×
“The convex optimization (28) was implemented in MATLAB using the fmincon interior-point method (which is itself an implementation of KNITRO , a high-quality interior-point trust-region method for large-scale nonlinear programming [13], [14]).”
Method
Research works printed but not linked: 2
- golub, 1996, Matrix Computations
- huber, 2004, Robust Statistics
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
“on rotations [35], [211]) has enabled even stronger results.”
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-citationSE-Sync: A certifiably correct algorithm for synchronization over the special Euclidean group
Rosen, D. M., Carlone, L., Bandeira, A. S., & Leonard, J. J. (2019) · The International Journal of Robotics Research · doi:10.1177/0278364918784361 · Cited by 282
“edience comes at the expense of reliability, as their restriction to local search renders these methods vulnerable to convergence to suboptimal critical points, even for relatively small noise levels [15, 36]. This observation, together with the fact that suboptimal critical points usually correspond to egregiously wr…”
Self-citationActive SLAM for Mobile Robots With Area Coverage and Obstacle Avoidance
Chen, Y., Huang, S., & Fitch, R. (2020) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2019.2963439 · Cited by 130
“The problem for conv SO ( n ) has been defined [24] π where (cid:13) • (cid:13) F is the Frobenius norm.”
BackgroundConvex Global 3D Registration with Lagrangian Duality
Briales, J., & González-Jiménez, J. (2017) · 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.2017.595 · Cited by 118
“Numerous state-of-the-art SLAM (Simultaneous Localization and Mapping) solutions work with models that include plane and line primitives too [2, 47, 17, 43, 28, 19].”
BackgroundLagrangian duality in 3D SLAM: Verification techniques and optimal solutions
Carlone, L., Rosen, D. M., Calafiore, G., Leonard, J. J., & Dellaert, F. (2015) · 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353364 · Cited by 95
“Using this parametrization, each summand in the objective in (6) becomes (using that (cid:107) R F = (cid:107) R T F in the first expression): where , and ⊗ is the Kronecker product.”
MethodSelf-citationZhang, Y., Shi, P., & Li, J. (2024) · The Photogrammetric Record · doi:10.1111/phor.12497 · Cited by 121
Advances in Inference and Representation for Simultaneous Localization and Mapping
Rosen, D. M., Doherty, K., Espinoza, A. T., & Leonard, J. J. (2021) · Annual Review of Control Robotics and Autonomous Systems · doi:10.1146/annurev-control-072720-082553 · Cited by 106
“While the idea of applying convex relaxation in SLAM is not new [indeed, many well-known initialization techniques are based upon this strategy (26, 27, 28, 29, 30)], what distinguishes certifiably correct methods from prior work is that the relaxations they employ are exact(4) provided that the noise on the data Ỹ in…”
BackgroundMethodSelf-citationSimultaneous Localization And Mapping: Present, Future, and the Robust-Perception Age.Preprint
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I. D., & Leonard, J. J. (2016) · arXiv (Cornell University) · Cited by 90
Self-citationA Certifiably Globally Optimal Solution to the Non-minimal Relative Pose Problem
Briales, J., Kneip, L., & González-Jiménez, J. (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.2018.00023 · Cited by 74
“Even though various recurrent tools exist, such as relaxing non-convex constraints to its convex hull [36, 34, 19] or applying Lagrangian duality [5], there remains much engineering involved in the combination of these tools to come up with a good relaxation.”
BackgroundCartan-Sync: Fast and Global SE(d)-Synchronization
Briales, J., & González-Jiménez, J. (2017) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2017.2718661 · Cited by 76
“For those datasets which were not originally isotropic we applied the same bounding approximation with isotropic covariances employed in previous works [2], [7].”
BackgroundMethodRandom-Finite-Set-Based Distributed Multirobot SLAM
Gao, L., Battistelli, G., & Chisci, L. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.3001664 · Cited by 63
Robust synchronization in SO(3) and SE(3) via low-rank and sparse matrix decomposition
Arrigoni, F., Rossi, B., Fragneto, P., & Fusiello, A. (2018) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2018.08.001 · Cited by 62
“In (Rosen et al., 2015) the Special Euclidean group is relaxed to its convex hull, which has a semi-definite representation (Saunderson et al.”
BackgroundA Certifiably Correct Algorithm for Synchronization over the Special Euclidean Group
Rosen, D. M., Carlone, L., Bandeira, A. S., & Leonard, J. J. (2020) · Springer proceedings in advanced robotics · doi:10.1007/978-3-030-43089-4_5 · Cited by 79
Self-citationDynamic Covariance Estimation — A parameter free approach to robust Sensor Fusion
Pfeifer, T., Lange, S., & Protzel, P. (2017) · IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI) · doi:10.1109/mfi.2017.8170347 · Cited by 33
“Papers like [16], [17] address this issue but whether these techniques can be applied to DCS or the self-tuning m-estimators is unclear.”
BackgroundAgostinho, S., Gomes, J. P., & Bue, A. D. (2022) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-022-01133-8 · Cited by 40
“[35] provided an in-depth analysis on the properties of the convex hull enclosing rotationmatrices, and theworks [7, 8,18,33,34,43] showed that despite the non-convex nature of the SO(3) group, there exist successful relaxation strategies.”
BackgroundSynchronization Problems in Computer Vision with Closed-Form Solutions
Arrigoni, F., & Fusiello, A. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01240-x · Cited by 39
“…specific properties of O ( d ) and SE ( d ), namely the fact that the matrix constructed from the pairwise measures is positive semidefinite when considering the Orthogonal Group [99,116], and the property that the convex hull of the Special Euclidean Group admits a semidefinite representation [90].”
BackgroundMethodInfluentialRobust Incremental SLAM Under Constrained Optimization Formulation
Bai, F., Vidal‐Calleja, T., & Huang, S. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2794610 · Cited by 30
“The techniques can be used to obtain a good initial value for the optimizer that is sufficiently close to the global minimum of the original problem [7], [8].”
MethodInvariant smoothing on Lie Groups
Chauchat, P., Barrau, A., & Bonnabel, S. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8594068 · Cited by 32
“To solve it, most methods revolve around using the manifold structure of the state-space and custom retractions [21] or using the matrix structure to relax the problem [32]–[34].”
MethodFast global optimality verification in 3D SLAM
Briales, J., & González-Jiménez, J. (2016) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2016.7759681 · Cited by 30
“…techniques for bootstrapping the iterative optimization [5]–[7], increasing the basin of convergence [4], [8], [9], analysis of the local convergence, problem structure, and influencing factors [10]–[15] or different relaxations of the problem [16] aimed at getting near-optimal global solutions.”
MethodSparse Pose Graph Optimization in Cycle Space
Bai, F., Vidal‐Calleja, T., & Grisetti, G. (2021) · IEEE Transactions on Robotics · doi:10.1109/tro.2021.3050328 · Cited by 28
“Carlone and Rosen et al. showed that the cost function can be chosen convex by using a Frobenius norm with an isotropic noise model [15], [45]–[48].”
BackgroundMethodSynchronization over Cartan Motion Groups via Contraction
Özyeşil, O., Sharon, N., & Singer, A. (2018) · SIAM Journal on Applied Algebra and Geometry · doi:10.1137/16m1106055 · Cited by 25
“Relaxing the constraints of SE(3) by its convex hull was considered in [45].”
Camera Motion from Group Synchronization
Arrigoni, F., Fusiello, A., & Rossi, B. (2016) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2016.64 · Cited by 21
“motion averaging) [17, 47, 49, 38, 6, 4]; • Σ = SL(d) homography synchronization [40]; • Σ = Sd permutation synchronization [37].”
BackgroundMethodMangelson, J. G., Liu, J., Eustice, R. M., & Vasudevan, R. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794454 · Cited by 27
“[4] relaxes optimization over the special orthogonal group (SO(d)) to the convex hull of SO(d) which can be represented using convex semidefinite constraints.”
BackgroundRelax and Recover: Guaranteed Range-Only Continuous Localization
Pacholska, M., Dümbgen, F., & Scholefield, A. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.2970952 · Cited by 20
“Therefore, it is common to use a standard non-linear least-squares solver such as Levenberg-Marquardt [8], which can recover a local minimum of (1) [9].”
Certifying Polynomial Nonnegativity via Hyperbolic Optimization
Saunderson, J. (2019) · SIAM Journal on Applied Algebra and Geometry · doi:10.1137/19m1253551 · Cited by 18
“uch as a rank one tensor) from noisy linear measurements, represented by y. Problems as diverse as noisy low-rank tensor completion [BM16] and simultaneous localization and mapping (SLAM) in robotics [RDL15] can be expressed in this form. Other cases of (2), in which V\R f 1;1gn, capture combinatorial optimization pro…”
BackgroundReisdorf, P., Pfeifer, T., Breßler, J., Bauer, S., Weissig, P., Lange, S., Wanielik, G., & Protzel, P. (2016) · Cited by 17
A review of optimisation strategies used in simultaneous localisation and mapping
Huang, S. (2018) · Journal of Control and Decision · doi:10.1080/23307706.2018.1552207 · Cited by 16
“The first attempt along this direction is done by Liu et al. (2012) using semi-definite programming and later by Rosen et al. (2015).”
MethodGiménez, J., Amicarelli, A., Toibero, J. M., di Sciascio, F. A., & Carelli, R. O. (2018) · Machine Intelligence Research · doi:10.1007/s11633-017-1109-4 · Cited by 15
“e, local minimum or saddle point), rather than the globally optimal solution[36].”
Convex Optimisation for Inverse Kinematics
Yenamandra, T., Bernard, F., Wang, J., Mueller, F., & Theobalt, C. (2019) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2019.00043 · Cited by 17
“Other approaches that consider SDP approaches for problems involving rotations have been demonstrated for SLAM [30], pose-graph optimisation [12], or rotation averaging [5, 17, 37].”
BackgroundRotation Coordinate Descent for Fast Globally Optimal Rotation Averaging
Parra, Á., Chng, S.-F., Chin, T.-J., Eriksson, A. P., & Reid, I. D. (2021) · 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/cvpr46437.2021.00428 · Cited by 22
“In vision and robotics, rotation averaging plays a crucial role in SfM [21, 25, 24, 8, 7, 20, 43, 36] and visual SLAM [5, 29, 33, 26, 19], in particular for initialising bundle adjustment.”
BackgroundNovel Parameterization for Gauss–Newton Methods in 3-D Pose Graph Optimization
Nasiri, S., Hosseini, R., & Moradi, H. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.3034021 · Cited by 15
“But Carlone et al. [15], [16] and Rosen et al. [17] showed that state-of-the-art algorithms are entangled in local minima and fail to converge to the global optimum in some cases.”
BackgroundOptimal Initialization Strategies for Range-Only Trajectory Estimation
Goudar, A., Dümbgen, F., Barfoot, T. D., & Schoellig, A. P. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3354623 · Cited by 15
“Unlike previous works, which focus on certifying batch solutions [15] or requiring additional sensing modalities [3], [10], [16], we focus on computationally tractable SDP relaxations to obtain optimal initial pose and trajectories over a short time horizon using multiple tags on the robot.”
BackgroundInitialization of 3D pose graph optimization using Lagrangian duality
Briales, J., & González-Jiménez, J. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989600 · Cited by 12
SCORE: A Second-Order Conic Initialization for Range-Aided SLAM
Papalia, A., Morales, J., Doherty, K., Rosen, D. M., & Leonard, J. J. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10160787 · Cited by 14
“unsurprising, as it is found in similar SLAM relaxations [7], [54] and the zeros elements indeed minimizes the problem cost.”
BackgroundSelf-citationA Linear Least Square Initialization Method for 3D Pose Graph Optimization Problem
Nasiri, S., Moradi, H., & Hosseini, R. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8460741 · Cited by 9
Comparison of two different objective functions in 2D point feature SLAM
Wang, H., Huang, S., Yang, G., & Dissanayake, G. (2018) · Automatica · doi:10.1016/j.automatica.2018.08.009 · Cited by 7
“…pose r 0 , that is, The state vector contains all the robot poses and all the feature positions that are to be estimated. x In (Carlone et al. 2016)(Rosen et al. 2015), the Frobenius norm is used to deal with the orientation part of odometry in (3), which generate the following objective function…”
BackgroundMethodInfluentialContinuous Probabilistic SLAM Solved via Iterated Conditional Modes
Giménez, J., Amicarelli, A., Toibero, J. M., di Sciascio, F. A., & Carelli, R. (2019) · Machine Intelligence Research · doi:10.1007/s11633-019-1186-7 · Cited by 7
Convex Optimisation for Inverse Kinematics
Yenamandra, T., Bernard, F., Wang, J., Mueller, F., & Theobalt, C. (2019) · MPG.PuRe (Max Planck Society) · Cited by 6
Incremental SQP method for constrained optimization formulation in SLAM
Bai, F., Huang, S., Vidal‐Calleja, T., & Zhang, Q. (2016) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv.2016.7838759 · Cited by 7
“Other methods are based on bootstrapping [7], [13], [14] exploiting the theory that an iterative optimization algorithm (for example Gauss-Newton) is likely to converge to the global optimum if the algorithm starts from a good initial estimate.”
MethodNon-convex Pose Graph Optimization in SLAM via Proximal Linearized Riemannian ADMM
Chen, X. C. X., Cui, C., Han, D., & Qi, L. (2025) · Journal of Optimization Theory and Applications · doi:10.1007/s10957-025-02759-5 · Cited by 11
“The function δ ( · ) is the indicator function of U which is defined as The iterative scheme of classical ADMM is given by In the above algorithm, the optimization subproblem (6b) and (6c) can be solved efficiently.”
Synchronization of Group-labelled Multi-graphs
Cin, A. P. D., Magri, L., Arrigoni, F., Fusiello, A., & Boracchi, G. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00639 · Cited by 7
“Existing techniques include spectral decomposition [9, 7], Lie-group optimization [29], diffusion over dual quaternions [54], Riemannian optimization [55], semidefinite programming [44, 43], distributed optimization [53], group contraction [39], Bayesian optimization [14] and deep learning [33, 27].”
MethodIncremental Cycle Bases for Cycle-Based Pose Graph Optimization
Forsgren, B., Brink, K., Ganesh, P., & McLain, T. W. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3236580 · Cited by 6
“Various techniques have been developed to increase the robustness of SLAM algorithms including obtaining a good initialization [4], [5], using robust cost functions [6], [7], using convex relaxations [8], [9], [10], and finding globally optimal solutions [11], [12].”
MethodFast Bayesian graph update for SLAM
Taguchi, S., Deguchi, H., Hirose, N., & Kidono, K. (2022) · Advanced Robotics · doi:10.1080/01691864.2021.2013939 · Cited by 5
“However, recent work [25,26] has shown that iterative techniques such as g2o fail to converge to an accurate estimate even in the case of fairly simple (real and simulated) 3D problems.”
BackgroundWang, H., Zhao, X., Huang, S., Li, Q., & Liu, Y. (2024) · Automatica · doi:10.1016/j.automatica.2024.111654 · Cited by 7
Satellite-Based Emitter Geolocation via Second-Order Cone Programming Initialization
Hao, C., Wan, X., Xu, X., & Yi, J. (2024) · IEEE Communications Letters · doi:10.1109/lcomm.2024.3435083 · Cited by 4
“Although these methods do not require initial values, they often exhibit limitations in geolocation accuracy [12].”
BackgroundCommunication Constrained Trajectory Alignment For Multi-Agent Inspection via Linear Programming
Mangelson, J. G., Vasudevan, R., & Eustice, R. M. (2018) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans.2018.8604775 · Cited by 4
“A variety of recent papers have investigated the use of the convex hull of SO(d) [18, 19]:”
BackgroundInvariant Smoothing on Lie Groups
Chauchat, P., Barrau, A., & Bonnabel, S. (2018) · HAL · doi:10.48550/arxiv.1803.02076 · Cited by 2
“To solve it, most methods revolve around using the manifold structure of the state-space and custom retractions [21] or using the matrix structure to relax the problem [32]–[34].”
MethodAdvances in Inference and Representation for Simultaneous Localization and MappingPreprint
Rosen, D. M., Doherty, K., Espinoza, A. T., & Leonard, J. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.05041 · Cited by 2
“While the idea of applying convex relaxation in SLAM is not new [indeed, many well-known initialization techniques are based upon this strategy (26, 27, 28, 29, 30)], what distinguishes certifiably correct methods from prior work is that the relaxations they employ are exact(4) provided that the noise on the data Ỹ in…”
BackgroundMethodSelf-citationFlorez, J.-D., Dor, M., & Tsiotras, P. (2025) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc63710.2025.11108108 · Cited by 2
Synchronization of Projective Transformations
Madhavan, R., Fusiello, A., & Arrigoni, F. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72913-3_2 · Cited by 2
“…most works propose proper extensions of the methodologies and algorithms originally developed for rotations: spectral decomposition [3,8,24,31,66]; Lie-group optimization [11,26]; iterative update based on neighbors [62]; semide(cid:28)nite programming [32,52,53,69]; modelling uncertainty [15,63].”
Wang, Y., & Peng, X. (2022) · Lecture notes in electrical engineering · doi:10.1007/978-981-16-9423-3_39 · Cited by 1
Visual metric and semantic localization for UGV
Zhang, H. (2022) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/162442
Expanding the reach of quantum optimization with fermionic embeddings
Zhao, A., & Rubin, N. C. (2024) · Quantum · doi:10.22331/q-2024-08-28-1451 · Cited by 2
Lipschitz Based Lower Bound Construction for Surrogate OptimizationPreprint
Almasi, M., Anahideh, H., & Rosenberger, J. M. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5229190
Pose Averaging from Relative Translations using a Riemannian Staircase
Fontana, E., & Tron, R. (2025) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc63710.2025.11107718 · Cited by 1
“One goal of SE-Sync is to find a soft spot between the rapidity of local search [12], [13], other previous approaches through constraint relaxations [2], [8], [12], [18], [22], [23], while still being able to offer some sort of certification of the algorithm’s output.”
Estimating lower bound functions in surrogate optimization
Almasi, M., Anahideh, H., & Rosenberger, J. M. (2026) · Computational Optimization and Applications · doi:10.1007/s10589-026-00824-x · Cited by 1
“Various techniques have been developed for constructing lower bounds on the optimal objective value, such as convex relaxation [35, 36], interval analysis [37, 38], statistical inference [39], and Lipschitz continuity [40].”
Metadata
- Authors
- Rosen, D. M., DuHadway, C., & Leonard, J. J.
- Affiliations
- MIT Computer Science and Artificial Intelligence Laboratory (US)
- Google (United States) (US)
- Venue
- 2015 IEEE International Conference on Robotics and Automation (ICRA)
- Publisher
- IEEE
- Published
- 2015-05
- Pages
- 5822-5829
- ISSN
- 1050-4729 (ISSN-L)
- OpenAlex
- W1553966841
- Cited by
- 56
- References
- 21
- Citations by year
- 2026: 1 · 2025: 4 · 2024: 6 · 2023: 2 · 2022: 4 · 2021: 5 · 2020: 5 · 2019: 6 · 2018: 11 · 2017: 4 · 2016: 7 · 2015: 1
- FWCI
- 13.27
- Type
- conference paper
- Language
- en
- Open access
- green
- hdl.handle.net
- Licence
- cc-by-nc-sa
- Locations
- Proceedings - IEEE International Conference on Robotics and Automation/Proceedings (publishedVersion)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc-sa)
- Funders
- Office of Naval Research
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Mathematics › Numerical Analysis › Advanced Optimization Algorithms Research
- Physical Sciences › Computer Science › Artificial Intelligence › Target Tracking and Data Fusion in Sensor Networks
- Keywords
- SLAM, convex relaxation, sparse inference, robot localization
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.