Iterated Kalman Filter Research Articles

Speed-sensorless control system of a bearingless induction motor based on iterative central difference Kalman filter

ABSTRACT In order to obtain speed self-detecting with low cost for a bearingless induction motor (BIM) a speed-sensorless control strategy based on the iterative central difference Kalman filter (ICDKF) is proposed. Firstly, on the basis of the BIM mathematical model, the nonlinear state equation is established and its order is reduced from fifth-order to fourth-order using the stator terminal voltage and current as input. Then, a sterling interpolation formulation is used in the filter to reduce the model error, and an iteration loop link is adopted to improve the filter accuracy. Finally, the online speed of the BIM is identified through the filter rotor speed estimation. Theoretical analysis, simulation and experimental results by UKF and CDKF method have been compared. The results show that the proposed speed-sensorless control system not only has good speed tracking performance and reduce the load disturbance but also improves the BIM suspension performance.

An Iterative Kalman Filter with Reduced-Biased Kalman Gain for Single Channel Speech Enhancement in Non-stationary Noise Condition

This paper presents an iterative Kalman filter (IT-KF) with a reduced-biased Kalman gain for single channel speech enhancement in Non-stationary Noise Conditions (NNCs). The proposed IT-KF aims to offset the bias in Kalman gain through efficient parameter estimation leading to improve the speech enhancement performance. To do this, we introduce a Decision Directed (DD) and a posteriori SNR based noise variance estimation method controlled through Speech Activity Detector (SAD). The proposed SAD incorporates a majority voting of three distinct SAD fusions. The LPC parameters are computed from the pre-smoothing of noisy speech. With these initial estimated parameters, an IT-KF processes the noisy speech at first iteration. The parameters are re-estimated from the processed speech, re-adjust the Kalman gain, and the process is repeated at second iteration. It is shown that the adjusted Kalman gain enables the IT-KF to minimize the remaining artifacts of the processed speech, yielding the enhanced speech. Extensive simulation results reveal that the proposed method outperforms other benchmark methods in NNCs for a wide range of SNRs.

Iterative Conditional Entropy Kalman filter (ICEKF) for noise reduction and Neuro Optimized Emotional Classifier (NOEC)

Emotion has a most important aspect in terms of interactions among the humans and this would become ideal for human emotions to get mechanically identified by the machines and primarily for enhancing the communication among the human–machine. In the recent work Enhanced Bat algorithm with Simulated Annealing (EBSA) are introduced for solving emotion recognition problem. Here the removal of noises from the speech samples and reduction in the number of speech features becomes very difficult task which reduces the accuracy of the classifier. To solve this problem this research work involves detection of emotions from speech which stimulates machines understanding human behavioral tasks namely reasoning, decision making and interaction. EBSA is used in the previous system to identify the happy, sad and neutral emotions from speech input. The performance of the previous system has been decreased due to recognition accuracy and feature selection. Improved Artificial Bee Colony (IABC) with Neuro Optimized Emotional Classifier (NOEC) solves this issue in the proposed system. The Iterative Conditional Entropy Kalman filtering (ICEKF) is initially processed to effectively filter the noisy features from the inputted speech data. Mel Frequency Cepstrum Coefficient (MFCC), pitch, energy, intensity and formants are extracted as speech features. Every extracted feature is maintained in the database and annotated along with their emotional class label. IABC algorithm chooses the feature optimally, which in turn employs the best fitness function values. From the optimally selected dataset, the NOEC is processed. Emotions can be identified from the Tamil news speech dataset with the help of the supervised machine learning technique, which demands the training set (collection of emotional speech recordings). Every recording or sample in the dataset is named with the emotional class and they are indicated as n-dimensional vector of spectrum coefficients which in turn is extracted from the Tamil news speech dataset. This dataset is collected from real time via using the search engine sites like Google, YouTube, twitter etc. By implementing IABC with NOEC classification process, the work segregates the emotional classes such as happy, sad, anger, fear and neutral emotions perfectly. From the experimental verification, it is confirmed that the proposed method IABC with NOEC gives better performances with respect to accuracy, precision, recall and f-measure values.

Regularized Extended Estimation With Stabilized Exponential Forgetting

This technical note concerns the problem of variable regularized estimation of time-varying nonlinear systems from the Bayesian viewpoint. The questions of how to impose the posterior information being variably regularized and how to forget this information are carefully discussed. The estimator design adopts the strategy of the iterated Kalman filter but differs in that, instead of the separated moments of the linearized system, only the augmented covariance matrix is updated. To suppress obsolete information, a decision problem involving the Kullback-Leibler divergence is solved. The decision provides the best combination of a pair of time-evolution model hypotheses in terms of the geometric mean. As a result, exponential forgetting with the adaptively tuned factor is inserted into the estimation process. The regularization of the investigated statistics is induced through the processing of externally supplied information. The presented estimator allows for absolute discarding or, conversely, retention of external information produced in terms of the Normal-Wishart distribution.

Generalized total Kalman filter algorithm of nonlinear dynamic errors-in-variables model with application on indoor mobile robot positioning

In this paper, a nonlinear dynamic errors-in-variables (DEIV) model which considers all of the random errors in both system equations and observation equations is presented. The nonlinear DEIV model is more general in the structure, which is an extension of the existing DEIV model. A generalized total Kalman filter (GTKF) algorithm that is capable of handling all of random errors in the respective equations of the nonlinear DEIV model is proposed based on the Gauss–Newton method. In addition, an approximate precision estimator of the posteriori state vector is derived. A two dimensional simulation experiment of indoor mobile robot positioning shows that the GTKF algorithm is statistically superior to the extended Kalman filter algorithm and the iterative Kalman filter (IKF) algorithm in terms of state estimation. Under the experimental conditions, the improvement rates of state variables of positions x, y and azimuth ψ of the GTKF algorithm are about 14, 29, and 66%, respectively, compared with the IKF algorithm.

A probabilistic collocation based iterative Kalman filter for landfill data assimilation

Abstract Accurate forecast of landfill gas (LFG) transport has remained as an active research area, due to the safety and environmental concerns, as well as the green energy potential. The iterative ensemble Kalman filter (IEnKF) has been used to characterize the heterogeneous permeability field of landfills. As a Monte Carlo-based method, IEnKF requires a sufficiently large ensemble size to guarantee its accuracy, which may result in a huge computational cost, especially for large-scale problems. In this study, an efficient probabilistic collocation based iterative Kalman filter (PCIKF) is developed. The polynomial chaos expansion (PCE) is employed to represent and propagate the uncertainties, and an iterative form of Kalman filter is used to assimilate the measurements. To further reduce the computational cost, only the zeroth and first-order ANOVA (analysis of variance) components are kept in the PCE approximation. As demonstrated by two numerical case studies, PCIKF shows significant superiority over IEnKF in terms of accuracy and efficiency. The developed method has the potential to reliably predict and develop best management practices for landfill gas production.

Iterative Solution of Relative Localization for Cooperative Multi-robot Using IEKF

Recently, multi-robot systems are emerging in various industries. The multi-robot systems have several advantages in comparison with a single-robot system. The task of single-robot system is limited, because a bigger robot is required to perform more multiple functions. However, multi-robot systems can distribute the functions to each robot. The localization problem is essential for realization and it becomes more important in the multi-robot system. The multi-robot system needs to work maintaining a formation. In order to accomplish the given mission in the demanded formation, the localization is important. For multi-robot localization, we use the relative position to find robots' position based on odometry sensor, and the iterative Kalman filter algorithm is utilized to estimate the accurate position.

Speech Enhancement Using Iterative Kalman Filter with Time and Frequency Mask in Different Noisy Environment

The main aim of the Speech Enhancement algorithms is to improve the Quality of speech. The Quality of speech is expressed in two parameters. One is clarity, and another is intelligibility. In this paper, we proposed a method to improve the quality of speech based on computationally efficient AR modeled Iterative Kalman Filter with time and frequency mask. This approach is based on reconstruction of noisy speech signals using Auto Regressive modeled Kalman filter and further to reduce artifact noise time and frequency mask is applied to the Kalman filter output. The results of the proposed method are found to be better compared to spectral subtraction, wiener filter and Kalman filter methods.

Structural identification based on incomplete measurements with iterative Kalman filter

Structural parameter evaluation and external force estimation are two important parts of structural health monitoring. But the structural parameter identification with limited input information is still a challenging problem. A new simultaneous identification method in time domain is proposed in this study to identify the structural parameters and evaluate the external force. Each sampling point in the time history of external force is taken as the unknowns in force evaluation. To reduce the number of unknowns for force evaluation the time domain measurements are divided into several windows. In each time window the structural excitation is decomposed by orthogonal polynomials. The time-variant excitation can be represented approximately by the linear combination of these orthogonal bases. Structural parameters and the coefficients of decomposition are added to the state variable to be identified. The extended Kalman filter (EKF) is augmented and selected as the mathematical tool for the implementation of state variable evaluation. The proposed method is validated numerically with simulation studies of a time-invariant linear structure, a hysteretic nonlinear structure and a time-variant linear shear frame, respectively. Results from the simulation studies indicate that the proposed method is capable of identifying the dynamic load and structural parameters fairly accurately. This method could also identify the time-variant and nonlinear structural parameter even with contaminated incomplete measurement.

Strap-down Inertial Navigation System Initial Alignment Based on Iterative Square-Root Cubature Kalman Filter

This paper presents an improved Cubature Kalman filter (CKF) for strap-down inertial navigation system (SINS) initial alignment. In order to enhance the stability and precision of the filter, Newton iteration algorithm and square-root strategy are blended into CKF. Iterative Square-root Cubature Kalman Filter (ISRCKF)makes full use of latest measurement information and directly employs the square root of covariance matrix for recursive update in the iterative process. ISRCKF can effectively avoid non-positive covariance matrix and improve the estimation accuracy. Simulation results show that ISRCKF performs well for SINS initial alignment with large azimuth misalignment.

Are central bank preferences asymmetric when policy targets vary over time?

The purpose of this paper is to analyze the effect of time varying monetary policy targets on the asymmetric preferences hypothesis for US monetary policy. Recent literature suggests that monetary policy responds asymmetrically to fluctuations in either an output gap or unemployment gap. Most of these studies impose the assumption of constant inflation and interest rate targets. This paper models both of these target rates as time varying parameters using a nested specification to test for constancy in the target rates. Additionally, the paper examines the estimation strategy needed to estimate all of the policy maker’s structural or deep parameters for the asymmetric preferences model. The model is estimated via maximum likelihood using an iterative Kalman filter. Results show that asymmetric policy response over the output gap disappears for all sample periods when the joint underlying dynamics of inflation and interest rate data are accounted for. Additionally, the results indicate that policy target rates are not well represented by constants for all sample periods examined. As a whole, the empirical exercise suggests that conclusions about monetary policy behavior might be sensitive to modeling assumptions about target policy rates.

Generalized Iterated Kalman Filter and its Performance Evaluation

In this paper, we present a generalized iterated Kalman filter (GIKF) algorithm for state estimation of a nonlinear stochastic discrete-time system with state-dependent multiplicative observation noise. The GIKF algorithm adopts the Newton-Raphson iterative optimization steps to yield an approximate maximum a posteriori estimate of the states. The mean-square estimation error (MSE) and the Cramér-Rao lower bound (CRLB) of the state estimates are also derived. In particular, the local convergence of MSE of GIKF is rigorously established. It is also proved that the GIKF yields a smaller MSE than those of the generalized extended Kalman filter and the traditional extended Kalman filter. The performance advantages and convergence of GIKF are demonstrated using Monte Carlo simulations on a target tracking application in a range measuring sensor network.

Navigation solution for a geostationary satellite based on its dynamic equations and occasional GNSS measurements

The paper offers solution to the problem of geostationary spacecraft navigation by navigation satellite signal measurements using nonlinear dynamic equations, which include perturbing factors such as non-sphericity of the gravitational field of the Earth, gravitational attraction of the Moon and the Sun, and direct solar radiation pressure. This solution uses the iterative Kalman filter based on UD-decomposition and is focused on the onboard implementation with limited computing resources. The estimates of current coordinates and velocity components obtained with the Kalman filter are taken into account when integrating nonlinear equations of a spacecraft motion up to the next solution time moment. With undetermined solar radiation pressure coefficient, the proposed solution proved efficient in terms of accuracy requirements for the spaceborne satellite navigation receiver. The tests were conducted using a GPS signal simulator and a receiver.

Trajectory-based visual localization in underwater surveying missions.

We present a new vision-based localization system applied to an autonomous underwater vehicle (AUV) with limited sensing and computation capabilities. The traditional EKF-SLAM approaches are usually expensive in terms of execution time; the approach presented in this paper strengthens this method by adopting a trajectory-based schema that reduces the computational requirements. The pose of the vehicle is estimated using an extended Kalman filter (EKF), which predicts the vehicle motion by means of a visual odometer and corrects these predictions using the data associations (loop closures) between the current frame and the previous ones. One of the most important steps in this procedure is the image registration method, as it reinforces the data association and, thus, makes it possible to close loops reliably. Since the use of standard EKFs entail linearization errors that can distort the vehicle pose estimations, the approach has also been tested using an iterated Kalman filter (IEKF). Experiments have been conducted using a real underwater vehicle in controlled scenarios and in shallow sea waters, showing an excellent performance with very small errors, both in the vehicle pose and in the overall trajectory estimates.

Performance enhancement for a GPS vector-tracking loop utilizing an adaptive iterated extended Kalman filter.

This paper deals with the problem of state estimation for the vector-tracking loop of a software-defined Global Positioning System (GPS) receiver. For a nonlinear system that has the model error and white Gaussian noise, a noise statistics estimator is used to estimate the model error, and based on this, a modified iterated extended Kalman filter (IEKF) named adaptive iterated Kalman filter (AIEKF) is proposed. A vector-tracking GPS receiver utilizing AIEKF is implemented to evaluate the performance of the proposed method. Through road tests, it is shown that the proposed method has an obvious accuracy advantage over the IEKF and Adaptive Extended Kalman filter (AEKF) in position determination. The results show that the proposed method is effective to reduce the root-mean-square error (RMSE) of position (including longitude, latitude and altitude). Comparing with EKF, the position RMSE values of AIEKF are reduced by about 45.1%, 40.9% and 54.6% in the east, north and up directions, respectively. Comparing with IEKF, the position RMSE values of AIEKF are reduced by about 25.7%, 19.3% and 35.7% in the east, north and up directions, respectively. Compared with AEKF, the position RMSE values of AIEKF are reduced by about 21.6%, 15.5% and 30.7% in the east, north and up directions, respectively.

Bond Risk Premia Forecasting: A Simple Approach for Extracting Macroeconomic Information from a Panel of Indicators

We propose a simple but effective estimation procedure to extract the level and the volatility dynamics of a latent macroeconomic factor from a panel of observable indicators. Our approach is based on a multivariate conditionally heteroskedastic exact factor model that can take into account the heteroskedasticity feature shown by most macroeconomic variables and relies on an iterated Kalman filter procedure. In simulations we show the unbiasedness of the proposed estimator and its superiority to different approaches introduced in the literature. Simulation results are confirmed in applications to real inflation data with the goal of forecasting long-term bond risk premia. Moreover, we find that the extracted level and conditional variance of the latent factor for inflation are strongly related to NBER business cycles.

An adaptive sparse-grid iterative ensemble Kalman filter approach for parameter field estimation

A new sparse-grid (SG) iterative ensemble Kalman filter (IEnKF) approach is proposed for estimating spatially varying parameters. The adaptive high-order hierarchical sparse-grid (aHHSG) method is adopted to discretize the unknown parameter field. An IEnKF is used to explore the parameter space and estimate the surpluses of the aHHSG interpolant at each SG level. Moreover, the estimated aHHSG interpolant on coarser levels is employed to provide a good initial guess of the IEnKF solver for the approximation on the finer levels. The method is demonstrated in estimating permeability field in flows through porous media.

Study on Signal De-Noising for Vehicle Active Anti-Collision Warning Radar

The Vehicle Active Anti-collision Warning system often adopts the millimeter-wave radar with linear frequency-modulated continuous (FMCW) system as its signal acquisition and processing device. Affected by the road environment and their own devices, intermediate frequency (IF) signal inevitably exist in a variety of interference and noise. This paper proposed an iterative Kalman filter algorithm to remove the noise from IF signal. And used of MATLAB to do the simulation analysis. The analysis results show that the improved algorithm enhances the precision of de-noising effectively and meet the requirements of real-time and accuracy.

Improved precise point positioning in the presence of ionospheric scintillation

Ionospheric disturbances can be detrimental to accuracy and reliability of GNSS positioning. We focus on how ionospheric scintillation induces significant degradation to Precise Point Positioning (PPP) and how to improve the performance of PPP during ionospheric scintillation periods. We briefly describe these problems and give the physical explanation of highly correlated phenomenon of degraded PPP estimates and occurrence of ionospheric scintillation. Three possible reasons can contribute to significant accuracy degradation in the presence of ionospheric scintillation: (a) unexpected loss of lock of tracked satellites which greatly reduces the available observations and considerably weakens the geometry, (b) abnormal blunders which are not properly mitigated by positioning programs, and (c) failure of cycle slip detection algorithms due to the high rate of total electronic content. The latter two reasons are confirmed as the major causes of sudden accuracy degradation by means of a comparative analysis. To reduce their adverse effect on positioning, an improved approach based on a robust iterative Kalman filter is adopted to enhance the PPP performance. Before the data enter the filter, the differential code biases are used for GNSS data quality checking. Any satellite whose C1---P1 and P1---P2 biases exceed 10 and 30 m, respectively, will be rejected. Both the Melbourne---Wubbena and geometry-free combination are used for cycle slip detection. But the thresholds are set more flexibly when ionospheric conditions become unusual. With these steps, most of the outliers and cycle slips can be effectively detected, and a first PPP estimation can be carried out. Furthermore, an iterative PPP estimator is utilized to mitigate the remaining gross errors and cycle slips which will be reflected in the posterior residuals. Further validation tests based on extensive experiments confirm our physical explanation and the new approach. The results show that the improved approach effectively avoids a large number of ambiguity resets which would otherwise be necessary. It reduces the number of re-parameterized phase ambiguities by approximately half, without scarifying the accuracy and reliability of the PPP solution.

Iterated Cubature Kalman Filter for State Estimation of Maneuver Reentry Vehicle

We present the new filters named iterated cubature Kalman filter (ICKF). The ICKF is implemented easily and involves the iterate process for fully exploiting the latest measurement in the measurement update so as to achieve the high accuracy of state estimation We apply the ICKF to state estimation for maneuver reentry vehicle. Simulation results indicate ICKF outperforms over the unscented Kalman filter and square root cubature Kalman filter in state estimation accuracy.