Comparison and combination of EAKF and SIR-PF in the Bayesian filter framework

Bayesian estimation theory provides a general approach for the state estimate of linear or nonlinear and Gaussian or non-Gaussian systems. In this study, we first explore two Bayesian-based methods： ensemble adjustment Kalman filter（EAKF） and sequential importance resampling particle filter（SIR-PF）, using a well-known nonlinear and non-Gaussian model（Lorenz ＇63 model）. The EAKF, which is a deterministic scheme of the ensemble Kalman filter（En KF）, performs better than the classical（stochastic） En KF in a general framework. Comparison between the SIR-PF and the EAKF reveals that the former outperforms the latter if ensemble size is so large that can avoid the filter degeneracy, and vice versa. The impact of the probability density functions and effective ensemble sizes on assimilation performances are also explored. On the basis of comparisons between the SIR-PF and the EAKF, a mixture filter, called ensemble adjustment Kalman particle filter（EAKPF）, is proposed to combine their both merits. Similar to the ensemble Kalman particle filter, which combines the stochastic En KF and SIR-PF analysis schemes with a tuning parameter, the new mixture filter essentially provides a continuous interpolation between the EAKF and SIR-PF. The same Lorenz ＇63 model is used as a testbed, showing that the EAKPF is able to overcome filter degeneracy while maintaining the non-Gaussian nature, and performs better than the EAKF given limited ensemble size.

Relative-Position Estimation Based on Loosely Coupled UWB–IMU Fusion for Wearable IoT Devices

Relative positioning is one of the important techniques in collaborativerobotics, autonomous vehicles, and virtual/augmented reality (VR/AR)applications. Recently, ultra-wideband (UWB) has been utilized to calculaterelative position as it does not require a line of sight compared to a camerato calculate the range between two objects with centimeter-level accuracy.However, the single UWB range measurement cannot provide the relativeposition and attitude of any device in three dimensions (3D) because oflacking bearing information. In this paper, we have proposed a UWB-IMUfusion-based relative position system to provide accurate relative positionand attitude between wearable Internet of Things (IoT) devices in 3D. Weintroduce a distributed Euler angle antenna orientationwhich can be equippedwith the mobile structure to enable relative positioning. Moving average andmin-max removing preprocessing filters are introduced to reduce the standarddeviation. The standard multilateration method is modified to calculate therelative position between mobile structures. We combine UWB and IMUmeasurements in a probabilistic framework that enables users to calculatethe relative position between two nodes with less error. We have carried outdifferent experiments to illustrate the advantages of fusing IMU and UWBranges for relative positioning systems. We have achieved a mean accuracy of0.31m for 3D relative positioning in indoor line of sight conditions.

Smoother and Bayesian filter based semi-codeless tracking of dual-frequency GPS signals

To precisely determine the integrated orbit of the Chinese manned spacecraft mission, a smoother and Bayesian filter based technique for optimum semi-codeless tracking of the P（Y） code on dual-frequency GPS signals has been advanced. This signal processing technique has been proven effective and robust for affording access to dual-frequency GPS signals. This paper introduces the signal dynamics and measurement models, describes the W o D bit estimation method, and corrects the mistakes of direct estimation of W bit in current semi-codeless tracking. Median filter is chosen as a smoother to find the best measurements at the current time among the history and current information. The Bayesian filter is used to track the L2 P（Y） code phase and L2 carrier phase recursively.

Recursive Filter with Partial Knowledge on Inputs and Outputs

This paper investigates the problem of state estimation for discrete-time stochastic linear systems, where additional knowledge on the unknown inputs is available at an aggregate level and the knowledge on the missing measurements can be described by a known stochastic distribution. Firstly, the available knowledge on the unknown inputs and the state equation is used to form the prior distribution of the state vector at each time step. Secondly, to obtain an analytically tractable likelihood function, the effect of missing measurements is broken down into a systematic part and a random part, and the latter is modeled as part of the observation noise. Then, a recursive filter is obtained based on Bayesian inference. Finally, a numerical example is provided to evaluate the performance of the proposed methods.