Outlier-Robust Estimation: Hardness, Minimally Tuned Algorithms, and Applications

被引:32
|
作者
Antonante, Pasquale [1 ]
Tzoumas, Vasileios [1 ,2 ]
Yang, Heng [1 ]
Carlone, Luca [1 ]
机构
[1] MIT, Lab Informat & Decis Syst, Cambridge, MA 02139 USA
[2] Univ Michigan, Dept Aerosp Engn, Ann Arbor, MI 48109 USA
关键词
Estimation; Signal processing algorithms; Probabilistic logic; Approximation algorithms; Simultaneous localization and mapping; Particle measurements; Measurement uncertainty; Algorithms; autonomous systems; computational complexity; computer vision; maximum likelihood estimation; resilient perception; robust estimation; SET MAXIMIZATION; REGISTRATION; REJECTION; OPTIMIZATION; STATISTICS; PERCEPTION; CONSENSUS; MOTION;
D O I
10.1109/TRO.2021.3094984
中图分类号
TP24 [机器人技术];
学科分类号
080202 ; 1405 ;
摘要
Nonlinear estimation in robotics and vision is typically plagued with outliers due to wrong data association or incorrect detections from signal processing and machine learning methods. This article introduces two unifying formulations for outlier-robust estimation, generalized maximum consensus ($\textrm{G}$-$\textrm{MC}$) and generalized truncated least squares ($\textrm{G-TLS}$), and investigates fundamental limits, practical algorithms, and applications. Our first contribution is a proof that outlier-robust estimation is inapproximable: In the worst case, it is impossible to (even approximately) find the set of outliers, even with slower-than-polynomial-time algorithms (particularly, algorithms running in quasi-polynomial time). As a second contribution, we review and extend two general-purpose algorithms. The first, adaptive trimming ($\textrm{ADAPT}$), is combinatorial and is suitable for $\textrm{G}$-$\textrm{MC}$; the second, graduated nonconvexity ($\textrm{GNC}$), is based on homotopy methods and is suitable for $\textrm{G-TLS}$. We extend $\textrm{ADAPT}$ and $\textrm{GNC}$ to the case where the user does not have prior knowledge of the inlier-noise statistics (or the statistics may vary over time) and is unable to guess a reasonable threshold to separate inliers from outliers (as the one commonly used in RANdom SAmple Consensus $(\textrm{RANSAC})$. We propose the first minimally tuned algorithms for outlier rejection, which dynamically decide how to separate inliers from outliers. Our third contribution is an evaluation of the proposed algorithms on robot perception problems: mesh registration, image-based object detection (shape alignment), and pose graph optimization. $\textrm{ADAPT}$ and $\textrm{GNC}$ execute in real time, are deterministic, outperform $\textrm{RANSAC}$, and are robust up to 80-90% outliers. Their minimally tuned versions also compare favorably with the state of the art, even though they do not rely on a noise bound for the inliers.
引用
收藏
页码:281 / 301
页数:21
相关论文
共 50 条
  • [31] Outlier-Robust Subsampling Techniques for Persistent Homology
    Stolz, Bernadette J.
    JOURNAL OF MACHINE LEARNING RESEARCH, 2023, 24
  • [32] Outlier-Robust Calibration Method for Sensor Networks
    Dorffer, Clement
    Puigt, Matthieu
    Delmaire, Gilles
    Roussel, Gilles
    2017 IEEE INTERNATIONAL WORKSHOP OF ELECTRONICS, CONTROL, MEASUREMENT, SIGNALS AND THEIR APPLICATION TO MECHATRONICS (ECMSM), 2017,
  • [33] Outlier-Robust Bayesian Multinomial Choice Modeling
    Benoit, Dries F.
    Van Aelst, Stefan
    Van den Poel, Dirk
    JOURNAL OF APPLIED ECONOMETRICS, 2016, 31 (07) : 1445 - 1466
  • [34] Outlier-robust set-membership estimation for discrete-time linear systems
    Chen, Aijun
    Dinh, Thach Ngoc
    Raissi, Tarek
    Shen, Yi
    INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL, 2022, 32 (04) : 2313 - 2329
  • [35] Statistical Similarity Measure-Based Adaptive Outlier-Robust State Estimator With Applications
    Bai, Mingming
    Huang, Yulong
    Zhang, Yonggang
    Chambers, Jonathon
    IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2022, 67 (08) : 4354 - 4361
  • [36] A Fast Outlier-robust Fusion Estimator for Local Bus Frequency Estimation in Power Systems
    Farahani, Ali
    Abolmasoumi, Amir H.
    Bayat, Mohammad
    Mili, Lamine
    2020 10TH SMART GRID CONFERENCE (SGC), 2020,
  • [37] Anomaly-Aware Network Traffic Estimation via Outlier-Robust Tensor Completion
    Wang, Qianqian
    Chen, Lei
    Wang, Qin
    Zhu, Hongbo
    Wang, Xianbin
    IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 2020, 17 (04): : 2677 - 2689
  • [38] Outlier-robust methods for forecasting realized covariance matrices
    Li, Dan
    Drovandi, Christopher
    Clements, Adam
    INTERNATIONAL JOURNAL OF FORECASTING, 2024, 40 (01) : 392 - 408
  • [39] Outlier-Robust Gromov-Wasserstein for Graph Data
    Kong, Lemin
    Li, Jiajin
    Tang, Jianheng
    So, Anthony Man-Cho
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [40] Outlier-robust kalman filter in the presence of correlated measurements
    Wang, Hongwei
    Liu, Yuanyuan
    Zhang, Wei
    Zuo, Junyi
    SIGNAL PROCESSING, 2022, 193