OptimalAction

PURPOSE

Returns the optimal action for a given belief.

SYNOPSIS

function [a v]=OptimalAction(P,b)

DESCRIPTION

   Returns the optimal action for a given belief.

   Returns the action associated with the element that maximizes the Policy
   value for a given belief.
   Additionally it also returns the maximazing value for the belief.

   See also Policy.

CROSS-REFERENCE INFORMATION

This function calls:

Values Belief value with all alpha-elements in the Policy.
max Upper bound of a CSpace

This function is called by:

MakeFigure3 Generates Figure 3.
MakeFigure4 Generates Figure 4.
SimulateFrom Simulates a POMDP from a given belief.

SOURCE CODE

0001 function [a v]=OptimalAction(P,b)
0002 %   Returns the optimal action for a given belief.
0003 %
0004 %   Returns the action associated with the element that maximizes the Policy
0005 %   value for a given belief.
0006 %   Additionally it also returns the maximazing value for the belief.
0007 %
0008 %   See also Policy.
0009 
0010   [v nAlpha]=max(Values(P,b));
0011   a=P.actions{nAlpha};