GetObsModelFixedS

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

   Instantiates the observation model for a particular state 's'.

   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.

   Observe that the output vector is normalized. In this way we do not
   need to explicitly compute the normalization factor, p(s).

   See also CS_CO_ObsModel

0001 function p=GetObsModelFixedS(OM,s)
0002 %   Defines p(o|s) for a fixed 's'.
0003 %
0004 %   Instantiates the observation model for a particular state 's'.
0005 %
0006 %   The CS_CO_ObsModel defines p(o,s) via Kernel smoothing: using a double
0007 %   Gaussian on 's' and on 'o'. From here:
0008 %       p(o|s)=p(o,s)/p(s)
0009 %   Where we assume p(s) as uniform.
0010 %
0011 %   Observe that the output vector is normalized. In this way we do not
0012 %   need to explicitly compute the normalization factor, p(s).
0013 %
0014 %   See also CS_CO_ObsModel
0015 
0016   w=OM.w.*cellfun(@(x)(Value(x,s)),OM.gS);
0017   p=Normalize(GMixture(w,OM.gO));
0018