Path: Common/Estimation
% Unscented Kalman Filter weights.
The weight matrix is used by the matrix form of the Unscented
Transform. Both UKFPredict and UKFUpdate use the data structure
generated by this function.
The constant alpha determines the spread of the sigma points around x
and is usually set to between 1e-3 and 1. beta incorporates prior
knowledge of the distribution of x and is 2 for a Gaussian
distribution. kappa is set to 0 for state estimation and 3 - number of
states for parameter estimation.
--------------------------------------------------------------------------
Form:
d = UKFWeight( d )
--------------------------------------------------------------------------
------
Inputs
------
d (.) Data structure with constants
.kappa (1,1) 0 for state estimation, 3-#states for
parameter estimation
.m (:,1) Vector of mean states
.alpha (1,1) Determines spread of sigma points
.beta (1,1) Prior knowledge - 2 for Gaussian
-------
Outputs
-------
d (.) Data structure with updated weights
.w (2*n+1,2*n+1) Weight matrix
.wM (1,2*n+1) Weight array
.wC (1,2*n+1) Weight array for covariance
.c (1,1) Scaling constant
.lambda (1,1) Scaling constant
--------------------------------------------------------------------------
Math: Linear/DupVect
Back to the Common Module page