Nga-Viet Nguyen

Power system state estimation with fusion method

A distributed power system consisting of multiple subsystems with individual state estimators needs to be globallymonitored using a system-wide state estimation function. To solve this problem, researchers have used several methods which can be categorized into three approaches: integrated state estimation, parallel state estimation and distributed state estimation. In literature, these methods are often separately applied and their temporary failures or states of poor quality can degrade the global state estimation.We propose a fusion method being able to combine different state estimation solutions in order to obtain a more reliable and accurate system-wide state estimation

An Observation Model Based on Polyline Map for Autonomous Vehicle Localization

Solution of the localization problem for autonomous vehicle navigation is an urgent requirement. In the wake of this requirement a new map-based method for the localization of autonomous vehicles using the extended Kalman Alter (EKF) is proposed. Formulation of the EKF equations is based upon a 4-wheel vehicle equipped with encoders, laser rangefinder and a polyline map. The observation model is comprised of special scanned points. The equations are derived for both range and bearing to form an effective observation model for the EKF estimator. Once the matching is set up, the pose predicted by dead reckoning will be well corrected for a robust localization.

Robust algorithm in distributed estimation fusion with correlation of local estimates

In distributed estimation fusion, locally obtained estimates are transmitted to the central processor via noisy channels. Traditionally, optimal linear methods are applied to solve the fusion problem under Gaussian noise assumption that can be severely violated in practise when channel noises are heavy-tailed. Hence, those methods should be replaced by robust analogs. M-estimates are well-known robust tools; however, when there is considerable correlation between local estimates, fusion accuracy may decrease. Thus, we propose a robust fusion algorithm based on a procedure for trimming outliers and the subsequent application of an optimal fusion method. Numerical experiments show that the proposed method is more accurate than conventional M-estimates, especially when there is a high degree of correlation involved.

A robust two-stage multisensor fusion in contaminated Gaussian channel noise

To solve the problem of distributed multisensor fusion, conventional fusion methods can be efficiently used in Gaussian noise models. In practice, the channel noise is usually non-Gaussian making those methods fail. A robust two-stage fusion algorithm based on the preliminary rejection of outliers in the data with the subsequent application of the conventional fusion method to the rest of the data is proposed. This fusion algorithm exhibits both high robustness in heavy contaminated Gaussian channel noise and good efficiency in nearly Gaussian channel noise both with small and large numbers of sensors.

Trimmed robust fusion under non-gaussian channel noise

In distributed multisensor fusion, local estimates may have to be communicated to a distant central processor. Hence, the communication channel noise is an important factor to the fusion algorithm. Optimal linear methods can be applied when channel noise is supposed to be Gaussian. In practice, the channel noise is not Gaussian and usually modeled by a contaminated Gaussian distribution. A two-stage trimmed robust fusion (TRIMRF) algorithm is proposed to adapt an optimal method to this case. The advantages of this method are its simplicity and capability of working effectively with little assumption about the underlying channel noise distribution.

Comparison of multi-sensor fusion filters weighted by scalars and matrices

Two fusion formulas with scalar and matrix weights are presented. The statistical relationship between them is established. They are compared on the multi-sensor Kalman filtering problem. The basic equation for cross-covariance of the local Kalman estimates being the key quantity for distributed fusion is derived. Examples demonstrating the desirable accuracy of the proposed fusion filters are given.