Kalman Filter With Dynamical Setting of Optimal Process Noise Covariance

Basso, Gabriel F.; Amorim, Thulio; Brito, Alisson V.; Nascimento, Tiago P.

doi:10.1109/access.2017.2697072

Cited by 36 publications

(14 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[19] developed the tracking algorithm with Kalman filter, [20] introduced a Kalman filter to estimate the object state for further tracking the desired dynamic object and filtering the noise, [21] used an adaptive Kalman filter to construct the motion model in the tracking process. Dynamical setting Kalman filter [22] was proposed to dynamically set the optimal process error covariance matrix for a constant velocity model Kalman filter, which can track a real erratic object. In addition, adaptive Kalman filter was combined with mean shift [23] or Camshift [24] to improve the robustness.…”

Section: Kalman Filter In Trackingmentioning

confidence: 99%

Stably Adaptive Anti-Occlusion Siamese Region Proposal Network for Real-Time Object Tracking

Zhang

2020

IEEE Access

View full text Add to dashboard Cite

Siamese region proposal network has made remarkable achievements in visual object tracking because of its balanced accuracy and speed. However, it regards tracking as a local one-shot detection task, which lose the power of updating the appearance model online thereby cannot handle the object-occlusion, fast motion and out-of-view situations. To tackle this problem, we propose a method that combines adaptive Kalman filter with Siamese region proposal network (Anti-occlusion-SiamRPN) to make full use of the object spatial-temporal information. Specifically we first extract target features through deep network and then uses adaptive Kalman filter to predict target trajectory in these difficult scenarios. Further this trajectory is used to select the candidate area of the next frame for Siamese region proposal network, which improve the searching mechanism. In this way, the introduction of adaptive Kalman filter makes the tracking process online learning which makes up for the disadvantage that Siamese region proposal network can only track offline. In addition, a hard example discrimination method (HEDM) is proposed to estimate whether the occlusion occurs and how seriously it is, which also improve Kalman filtering mechanism to make it update adaptively. Our method being evaluated with the speed of 80 FPS on five widely-applied challenging benchmarks including OTB2013, OTB2015, OTB50, VOT2016 and VOT2018. The extensive experimental results demonstrate our method achieves state-of-the-art effects and great improvement in comparison to other trackers. INDEX TERMS Visual tracking, Siamese region proposal network, Kalman filter, occlusion judgment, real-time tracker.

show abstract

Section: Kalman Filter In Trackingmentioning

confidence: 99%

Stably Adaptive Anti-Occlusion Siamese Region Proposal Network for Real-Time Object Tracking

Zhang

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Improper choice of Q and R may influence the performance of the filter method. Many estimation methods have been proposed to determine Q and R in the filter process (e.g., Akhlaghi et al 2017;Basso et al 2017;Ding et al 2007;Liebich et al 2017;Saho and Masugi 2015). With the use of adaptive filtering approach, we estimate Q and R , and it can be expressed as (Akhlaghi et al 2017) where ε(k) = Y (k) − HX(k|k) is the difference between the observed value and estimated one.…”

Section: Adaptive Kalman Filtermentioning

confidence: 99%

A reduction method for magnetic disturbance caused by DC railway system

Ding

et al. 2021

Earth Planets Space

View full text Add to dashboard Cite

The stray current of direct current (DC) railway systems causes magnetic disturbance in geomagnetic measurements, which may complicate the identification of useful information. The magnetic disturbance exhibits broadband characteristics in the frequency domain. In this paper, we propose a noise reduction method based on the adaptive Kalman filter to extract useful signals from the geomagnetic data with a high level of noise. The covariance matrixes of both the process noise (Q) and measurement noise (R) can be adaptively estimated to improve the performance of the adaptive Kalman filter. The proposed method is adopted to process the geomagnetic data collected at the Beijing Geomagnetic Observatory (BJI), which is affected by the DC railway system. The magnetic disturbance is largely reduced, and the signal-to-noise ratios of the horizontal and vertical components of the geomagnetic field are improved by more than 14 dB and 27 dB, respectively. The K-indices are calculated to evaluate the performance of the adaptive Kalman filter method. To assess the influence of the adaptive Kalman filter on natural signals, the geomagnetic data that contain rapid variations are processed. The denoising results show that the adaptive Kalman filter can effectively reduce the magnetic disturbance caused by DC railway system without large impact on the natural geomagnetic rapid variations.

show abstract

“…It is mainly divided into two categories. The first category, which typically includes UKF, is also based on the accurate system model and it can further reduce the model linearization error [8]; and the second category is represented by predictive filter [19], [20], H ∞ filter [21], [22], adaptive filter [23] and multiple model adaptive estimation [24], which can reduce or suppress the uncertainty influence in the system model. Besides, in order to ensure algorithm stability, nonlinear observer [25] and STF are put forward.…”

Section: Related Workmentioning

confidence: 99%

Turbulence Error Modeling and Restriction for Satellite Attitude Determination System Based on Improved Maximum Correntropy Kalman Filter

et al. 2019

View full text Add to dashboard Cite

In the process of satellite attitude determination, satellites or sensors themselves often encounter a variety of turbulence influences due to the complexity of space environments. Such influences can lead to the mutation and non-Gaussian noises for the attitude determination system. To solve these problems, in this paper, we construct a unified error model for the turbulence influences, which is a non-Gaussian noise model, and propose an improved attitude filter method to restrict the turbulence noises and the system mutation to enhance attitude determination accuracy and robustness. The unified error model combined with jitters and vibrations in the actual process of satellite attitude determination is firstly designed. Then an Improved Adaptive Kalman filter (IAKF) based on both the Strong Tracking Filter (STF) and the Maximum Correntropy Filter (MCKF) is put forward. By using of the optimization principle with both of fading factor and Maximum Correntropy Criterion (MCC), this proposed filter algorithm can suppress the influences of system mutations and non-Gaussian noises at the same time. It can eliminate the system mutations and the turbulence errors, and achieve excellent robustness and the attitude determination accuracy for the nonlinear system. Extensive simulations of the proposed filter are conducted under the conditions of the Gaussian noises, system mutation with large outliers, non-Gaussian noise with turbulence noises, and both the mutation and non-Gaussian turbulence error. The results demonstrate that our filter outperforms the existing attitude filter algorithms significantly.

show abstract

Kalman Filter With Dynamical Setting of Optimal Process Noise Covariance

Cited by 36 publications

References 21 publications

Stably Adaptive Anti-Occlusion Siamese Region Proposal Network for Real-Time Object Tracking

Stably Adaptive Anti-Occlusion Siamese Region Proposal Network for Real-Time Object Tracking

A reduction method for magnetic disturbance caused by DC railway system

Turbulence Error Modeling and Restriction for Satellite Attitude Determination System Based on Improved Maximum Correntropy Kalman Filter

Contact Info

Product

Resources

About