Performance evaluation of iterated extended Kalman filter with variable step-length

Havlík, Jindřich; Straka, Ondřej

doi:10.1088/1742-6596/659/1/012022

Cited by 41 publications

(41 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The adaptive IEKF has automatically stopped the iterations when there is no change in the next iteration. Another modification research of IEKF is improving the IEKF with only one GN (Gauss-Newton) method iteration which is suitable for a control of the step-length iterations [11].…”

Section: A Related Workmentioning

confidence: 99%

“…The EKF algorithm has better performance in mildly nonlinear functions, but if the observation matrix equation (6) is strongly nonlinear (such as in tracking system), the performance of the EKF deteriorates. In such situations, the iterated EKF (IEKF) tends to provide more accurate estimates than EKF [11], [12], [19], [24]. IEKF algorithm can be improved the estimation output from EKF algorithm.…”

Section: Iterated Extended Kalman Filter (Iekf)mentioning

confidence: 99%

“…The update states of IEKF were also obtained from each result of EKF algorithm. The main state is Jacobian matrix H k which is derived as [11]: (15) The Kalman gain should be changed by its iteration as follow:…”

Section: ) Update Statementioning

confidence: 99%

“…extended Kalman filter (EKF) provide a feasible solution to mitigate propagation effects and therefore it improves the accuracy of mobile target tracking in the indoor wireless environment [4], [9]. The EKF performs well for systems with mildly nonlinear functions, but if the measurement data is strongly nonlinear (such as tracking system), the performance of the filter deteriorates [11]. In such situation, the iterated extended Kalman filter (IEKF) achieves better consistency and accuracy compared to the EKF [1]1, [12].…”

Section: Introductionmentioning

confidence: 99%

“…The idea of IEKF is to improve the estimation by local iteration or forward data processing from the previous output estimation that is EKF algorithm. The iterations are usually stopped when there is no significant change in consecutive iterations or other criteria such as a maximum number iterations are met [11]. However, the time computation of IEKF is even slower than EKF, which limits its applications.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Modified Iterated Extended Kalman Filter for Mobile Cooperative Tracking System

Ainul

Kristalina

Sudarsono

2017

International Journal on Advanced Science, Engineering and Information Technology

View full text Add to dashboard Cite

Tracking a mobile node using wireless sensor network (WSN) under cooperative system among anchor node and the mobile node, has been discussed in this work, interested to the indoor positioning applications. Developing an indoor location tracking system based on received signal strength indicator (RSSI) of WSN is considered cost effective and the simplest method. The suitable technique for estimating position out of RSSI measurements is the extended Kalman filter (EKF) which is especially used for nonlinear data as RSSI. In order to reduce the estimated errors from EKF algorithm, this work adopted forward data processing of the EKF algorithm to improve the accuracy of the filtering output, its called iterated extended Kalman filter (IEKF). However, using IEKF algorithm should know the stopping criterion value that is influenced to the maximum number iterations of this system. The number of iterations performed will be affected to the computation time although it can improve the estimation position. In this paper, we propose modified IEKF for mobile cooperative tracking system within only 4 iterations number. The illustrated results using RSSI measurements and simulation in MATLAB show that our proposed method has the capability to reduce error estimation percentage up to 19.3% , with MSE (mean square error) 0.88 m compared with a conventional IEKF algorithm with MSE 1.09 m. The time computation performance of our proposed method achieved in 3.55 seconds which is better than adding more iteration process.

show abstract

Section: A Related Workmentioning

confidence: 99%

Section: Iterated Extended Kalman Filter (Iekf)mentioning

confidence: 99%

Section: ) Update Statementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Modified Iterated Extended Kalman Filter for Mobile Cooperative Tracking System

Ainul

Kristalina

Sudarsono

2017

International Journal on Advanced Science, Engineering and Information Technology

View full text Add to dashboard Cite

show abstract

Novel iterative cubature Kalman filters under maximum correntropy criterion for the robust state estimation

Lu,

Zhou,

et al. 2024

Asian Journal of Control

View full text Add to dashboard Cite

The cubature Kalman filter (CKF) based on maximum correntropy criterion (MCC) is robust under non‐Gaussian noises, but it may encounter numerical problems when there are large outliers. First, based on the MCC and weighted least square (WLS) methods, a new cost function is introduced to calculate the state and covariance updates, which can better solve numerical problems caused by large outliers. Then, a nonlinear measurement function is used, and the measurement information is updated through the latest iterative values, which can obtain more accurate results under various non‐Gaussian noises. In addition, the Gauss–Newton, Levenberg–Marquardt, and Quasi–Newton methods are used, and three iterative algorithms named GN‐IMCC‐CKF, LM‐IMCC‐CKF, and QN‐IMCC‐CKF are respectively derived for the updating steps. Compared with the existing algorithms, the simulation results of two classical models show that the proposed algorithms have better numerical stability and estimation accuracy under non‐Gaussian systems.

show abstract