A Robust Bayesian Approach for Online Filtering in the Presence of Contaminated Observations

Chughtai, Aamir Hussain; Tahir, Muhammad; Uppal, Momin

doi:10.1109/tim.2020.3033759

Cited by 9 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the same time, the model should remain amenable for VB inference. To this end, we choose the inference model from our previous work [30], with a few modifications. For a discrete time SSM, the process and measurement equations are given as follows…”

Section: Gross Error Modelingmentioning

confidence: 99%

“…For inferential tractability, the model as originally reported is simplified by ignoring the added randomness in the bias magnitude in (2). The bias evolution is expressed as follows where the modeling rationale remains the same as originally reported in [30].…”

Section: Gross Error Modelingmentioning

confidence: 99%

“…We assume m = 4, since for higher-dimensional problems even rejection-based methods like the SOR-UKF can have acceptable performance for a larger probability of errors owing to the redundancy of useful information in other uncorrupted dimensions. However, this does not limit the applicability of the proposed method for large m. Our point is to emphasize how properly utilizing information from an affected dimension, a characteristic of analytical redundancy approaches, is more useful in contrast to completely rejecting the information, a characteristic of hardware redundancy approaches [30]. In addition, we evaluate the relative performance of the proposed filter with similar analytical redundancy approaches.…”

Section: Numerical Experimentsmentioning

confidence: 99%

“…In [29], a similar model for describing bias is used. The use of random bias model has been proposed in [30] where the bias is represented with Gaussian density and subsequently the mean of the distribution is described in a Markovian manner. The use of Bernoulli random vectors is proposed to switch between multiple models catering for the presence and absence of gross errors.…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, the sole use of Gaussian approximations for inference no longer remains suitable. Therefore, the authors in [28], [29], [30], opt for the powerful PFs for the inference that can effectively handle arbitrary probability densities. The use of a large number of particles ranging from 15000 to 20000 is reported in these works.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Variational-Based Non-linear Bayesian Filtering with Observations Subjected to Gross Errors

Chughtai¹,

Majal²,

Tahir³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

State estimation of dynamical systems is crucial for providing new decision-making and system automation information in different applications. However, the assumptions on the standard computational models for sensor measurements can be violated in practice due to different types of data abnormalities such as random and gross errors. In this work, we focus on the occurrence of gross errors and propose a robust filter for their detection and mitigation during state estimation of nonlinear dynamical systems. We model the presence of gross errors within the generative structure of the state-space models. Subsequently, employing the theory of Variational Bayes and general Gaussian filtering, we devise a recursive filter which we call the Bias Detecting and Mitigating (BDM) filter. As the error detection mechanism is embedded within the filter structure its dependence on any external detector is obviated. Simulations verify the performance gains of the proposed BDM filter compared to similar Kalman filtering-based approaches in terms of robustness to temporary and persistent bias presence.

show abstract

Section: Gross Error Modelingmentioning

confidence: 99%

Section: Gross Error Modelingmentioning

confidence: 99%

Section: Numerical Experimentsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Variational-Based Non-linear Bayesian Filtering with Observations Subjected to Gross Errors

Chughtai¹,

Majal²,

Tahir³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

A variational Bayesian based robust filter for unknown measurement bias and inaccurate noise statistics

Yang,

Fu,

Zhang

2024

J. Inst.

View full text Add to dashboard Cite

In many practical fields, the unknown time-varying measurement biases (additive and multiplicative bias) and heavy-tailed measurement noise caused by some unpredictable anomalous behaviors may degrade the performance of conventional Kalman filter seriously. To solve the state estimation problem of systems with time-varying measurement biases and heavy-tailed measurement noise, this paper proposes a new variational Bayesian (VB) based robust filter. Firstly, the non-Gaussian measurement likelihood probability density function (ML-PDF) with multiplicative and additive measurement bias is built. Then, the conjugate prior distributions for unknown bias and noise scale parameters are selected, and the VB method is utilized to jointly infer the system state, unknown measurement biases and inaccurate measurement noise covariance matrix. Finally, a VB based robust filter is derived and its effectiveness is verified by the numerical simulations.

show abstract

Outlier-Robust Filtering for Nonlinear Systems With Selective Observations Rejection

2022

Self Cite

View full text Add to dashboard Cite

In this article, we consider the problem of outlierrobust state estimation where the measurement noise can be correlated. Outliers in data arise due to many reasons like sensor malfunctioning, environmental behaviors, communication glitches, etc. Moreover, noise correlation emerges in several realworld applications e.g. sensor networks, radar data, GPS-based systems, etc. We consider these effects in system modeling which is subsequently used for inference. We employ the Expectation-Maximization (EM) framework to derive both outlier-resilient filtering and smoothing methods, suitable for online and offline estimation respectively. The standard Gaussian filtering and the Gaussian Rauch-Tung-Striebel (RTS) smoothing results are leveraged to devise the estimators. In addition, Bayesian Cramer-Rao Bounds (BCRBs) for a filter and a smoother which can perfectly detect and reject outliers are presented. These serve as useful theoretical benchmarks to gauge the error performance of different estimators. Lastly, different numerical experiments, for an illustrative target tracking application, are carried out that indicate performance gains compared to similarly engineered state-of-the-art outlier-rejecting state estimators. The advantages are in terms of simpler implementation, enhanced estimation quality, and competitive computational performance.

show abstract

A Robust Bayesian Approach for Online Filtering in the Presence of Contaminated Observations

Cited by 9 publications

References 47 publications

Variational-Based Non-linear Bayesian Filtering with Observations Subjected to Gross Errors

Variational-Based Non-linear Bayesian Filtering with Observations Subjected to Gross Errors

A variational Bayesian based robust filter for unknown measurement bias and inaccurate noise statistics

Outlier-Robust Filtering for Nonlinear Systems With Selective Observations Rejection

Contact Info

Product

Resources

About