GetObsModelFixedO

PURPOSE

Defines p(o|s) for a fixed 'o'.

SYNOPSIS

function p=GetObsModelFixedO(OM,o)

DESCRIPTION

   Defines p(o|s) for a fixed 'o'.

   Instantiates the observation model for a particular observation 'o'.
   The CS_CO_ObsModel defines p(o,s) via Kernel smoothing: using a double
   Gaussian on 's' and on 'o'. From here:
       p(o|s)=p(o,s)/p(s)
   Where we assume p(s) as uniform.

   The output is not normalized.

   See also CS_CO_ObsModel

CROSS-REFERENCE INFORMATION

This function calls:

GMixture Gaussian mixture constructor.
Value Evaluates a GMixture.
Value Evaluation of a Gaussian.
UniformProbability Uniform probability value on a continuous space.
UniformProbability Uniform probability value on a discrete space.

This function is called by:

MakeFigure1 Generates Figure 1.
DiscretizeObsModel Discretizes the observation model on the observation side.
Update Belief evolution under an observation model.
ComputeAlpha_j_a_o Computes a particular alpha-element.
ComputeAlpha_j_a_o Computes a particular alpha-element.

SOURCE CODE

0001 function p=GetObsModelFixedO(OM,o)
0002 %   Defines p(o|s) for a fixed 'o'.
0003 %
0004 %   Instantiates the observation model for a particular observation 'o'.
0005 %   The CS_CO_ObsModel defines p(o,s) via Kernel smoothing: using a double
0006 %   Gaussian on 's' and on 'o'. From here:
0007 %       p(o|s)=p(o,s)/p(s)
0008 %   Where we assume p(s) as uniform.
0009 %
0010 %   The output is not normalized.
0011 %
0012 %   See also CS_CO_ObsModel
0013 
0014   w=OM.w.*cellfun(@(x)(Value(x,o)),OM.gO);
0015   ps=UniformProbability(OM.S);
0016   p=GMixture(w,OM.gS)/ps;
0017