This paper proposes a new adaptive extended Kalman filter (AEKF) for a class of nonlinear systems perturbed by noise which is not necessarily additive. The proposed filter is adaptive against the uncertainty in the process and measurement noise covariances. This is accomplished by deriving two recursive updating rules for the noise covariances, these rules are easy to implement and reduce the n...

In this thesis we focused on subsymbolic approach to machine game play problem. We worked on two different methods of learning. Our first goal was to test the ability of common feed-forward neural networks and the mixture of expert topology. We have derived reinforcement learning algorithm for mixture of expert network topology. This topology is capable to split the problem into smaller parts, ...

This paper deals with the problem of maneuvering target tracking in wireless tracking service. It results in a mixed linear/non-linear Models estimation problem. For maneuvering tracking systems, these problems are traditionally handled using the extended Kalman filter or Particle filter. In this paper, Marginalized Particle Filter is presented for applications in such problem. The algorithm ma...

In this paper, a novel application of state estimation in environmental engineering is presented. The objective is to use on-line estimation techniques including moving horizon estimator (MHE) and extended Kalman filer (EKF) for an early concentration estimation of toxic agent presented in water supply. These estimation techniques (MHE and EKF) can be integrated in an early warning system of se...

GPS/INS integrated system is very subject to uncertainties due to exogenous disturbances, device damage, and inaccurate sensor noise statistics. Conventional Kalman filer has no robustness to address system uncertainties which may corrupt filter performance and even cause filter divergence. Based on the INS error dynamic equation, a robust Kalman filter is analyzed and applied in loosely couple...

In this short paper, we describe the tractography method based on the intrinsic unscented Kalman filer (IUKF) [1]. This method is a generalization of unscented Kalman filter (UKF) and involves the use of intrinsic geometry of the space of symmetric positive definite matrices denoted henceforth by Pn. In this filter, operations that are intrinsic to Pn are employed and thus no explicit constrain...

in the several past years, extended kalman filter (ekf) and unscented kalman filter (ukf) havebecame basic algorithm for state-variables and parameters estimation of discrete nonlinear systems.the ukf has consistently outperformed for estimation. sometimes least estimation error doesn't yieldwith ukf for the most nonlinear systems. in this paper, we use a new approach for a two variablesta...

This paper presents an online sensor fusion algorithm for state estimation of a remotely operated underwater vehicle for aquaculture inspection. The algorithm is based on an Unscented Kalman Filer (UKF) and uses information from several sources including an onboard inertial sensor, an onboard camera combined with line lasers and a priory knowledge about the aquaculture geometry. The performance...

In this paper, we propose an approximate Bayesian computation approach to perform a multiple target tracking within a binary sensor network. The nature of the binary sensors (getting closer moving away information) do not allow the use of the classical tools (e.g. Kalman Filter, Particle Filer), because the exact likelihood is intractable. To overcome this, we use the particular feature of the ...

An Ensemble Kalman Filter (EnKF, the predictor) is used make a large change in the state, followed by a Particle Filer (PF, the corrector), which assigns importance weights to describe a non-Gaussian distribution. The importance weights are obtained by nonparametric density estimation. It is demonstrated on several numerical examples that the new predictor-corrector filter combines the advantag...