Code covered by the BSD License
-
ESS( w )
-
Gauss( x )
-
countPathHypotheses(x)
-
covParticlesXY( x,y,w )
-
h=error_ellipse(varargin)
ERROR_ELLIPSE - plot an error ellipse, or ellipsoid, defining confidence region
-
motionModel( X, action )
-
observationLikelihood( x, obs...
-
particleFilter( x,w, action, ...
A simple SIS with resampling filter
-
plotParticles( x, w )
Dibuja un conjunto de particulas de "poses" del robot
-
plotParticlesFromFile( f )
Fondo negro:
-
plotParticlesUnweight( x, w )
Dibuja un conjunto de particulas de "poses" del robot
-
resample( w )
-
resampleMultinomial( w )
-
resampleResidual( w )
"Repetition counts" (plus the random part, later on):
-
resampleStratified( w )
-
resampleSystematic( w )
-
runExample(INTERACTIVE,RESAMP...
The main script for running the Particle Filter example
-
runAnalysis.m
-
View all files
from
Resampling methods for particle filtering
by Jose-Luis Blanco
Implementation of four resampling methods (Multinomial, Residual, Stratified, and Systematic)
|
| covParticlesXY( x,y,w ) |
function [ COV, MEAN ] = covParticlesXY( x,y,w )
% ---------------------------------------------------------
% function [ COV ] = covParticlesXY( x,y,w )
%
% Computes the approximated 2x2 covariance matrix from a
% set of particles, given their "x" and "y" coordinates and their
% weights "w". It also computes the average "mean" value.
% Jose Luis Blanco Claraco, 26-JUN-2006
% ---------------------------------------------------------
% Assure weights are normalized:
w=w./sum(w);
% The mean values:
MEAN(1) = sum(x.*w);
MEAN(2) = sum(y.*w);
% The covariance:
var_x=0;
var_y=0;
var_xy=0;
var_x = sum( ((x - MEAN(1)).^2).*w );
var_y = sum( ((y - MEAN(2)).^2).*w );
var_xy = sum( ((x - MEAN(1)).*(y - MEAN(2))).*w );
COV(1,1)=var_x;
COV(2,2)=var_y;
COV(1,2)=var_xy;
COV(2,1)=var_xy;
|
|
Contact us at files@mathworks.com