GetTest1caParameters
PURPOSE 
The example on Figure 10.
SYNOPSIS
function [POMDP P]=GetTest1caParameters(varargin)
DESCRIPTION
The example on Figure 10.
This is the same example as GetTest1Parameters but with a continuous
action set.
Parameters
- nSampledActions: Samples to use when discretezing the actions
space.
Default value 4.
See also GetTest1Parameters.CROSS-REFERENCE INFORMATION
This function calls:
- CS_CA_ActionModel CS_CA_ActionModel constructor.
- get Get for GBeliefs.
- GetTest1Parameters The example on Figure 1.
- get Get function for the GMixture object.
- Gaussian Gaussian construtor.
- get Gaussian object get function.
- get Get function for CS_CO_CA_POMDPs.
- get Get function for CS_CO_DA_POMDPs.
- get Get function for CS_CO_POMDPs.
- CS_DO_CA_POMDP CS_DO_CA_POMDP constructor.
- get Get function for CS_DO_CA_POMDPs.
- get Get function for CS_DO_DA_POMDPs.
- get Get function for CS_POMDPs.
- get Get function for DS_CO_CA_POMDPs.
- get Get function for DS_CO_DA_POMDPs.
- get Get function for DS_DO_CA_POMDPs.
- get Get function for DS_DO_DA_POMDPs.
- POMDP POMDP constructor.
- get Get functio for POMDPs.
- CS_CA_RewardModel CS_CA_RewardModel constructor.
- CSpace CSpace constructor.
- dim Dimensionality of a continuous space.
- dim Dimensionality of a discrete space.
- GetTest1coParameters The example on Figure 11.
SOURCE CODE
0001 function [POMDP P]=GetTest1caParameters(varargin) 0002 % The example on Figure 10. 0003 % 0004 % This is the same example as GetTest1Parameters but with a continuous 0005 % action set. 0006 % 0007 % Parameters 0008 % - nSampledActions: Samples to use when discretezing the actions 0009 % space. 0010 % Default value 4. 0011 % 0012 % See also GetTest1Parameters. 0013 0014 % Test1 with the default paramters 0015 % ncBelief = 4 (not used) 0016 % ncAlpha = 9 0017 % actionScale = 2 0018 [POMDP P]=GetTest1Parameters(); 0019 0020 % Retrive information on the CS_DA_DO_POMDP 0021 gamma=get(POMDP,'gamma'); 0022 ncAlpha=get(POMDP,'maxAlphaC'); 0023 0024 S=get(POMDP,'StateSpace'); 0025 ds=dim(S); 0026 O=get(POMDP,'ObsSpace'); 0027 0028 OM=get(POMDP,'ObsModel'); 0029 0030 % Now define the Action CSpace, the CS_CA_ActionModel and the 0031 % CS_CA_RewardModel 0032 0033 %CSpace Action 0034 A=CSpace(-6,6); 0035 0036 % CS_CA_ActionModel 0037 B=eye(ds); % Assuming S and A have the same dimensionality 0038 Sa=0.05*eye(ds); 0039 AM=CS_CA_ActionModel(S,A,B,Sa); 0040 0041 % CS_CA_RewardModel 0042 w=[ -10 2 -10 ... 0043 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 ... 0044 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 ]; 0045 0046 gS={ Gaussian(-25,250) Gaussian(3,3) Gaussian(25,250) ... 0047 Gaussian(-21,1) Gaussian(-19,1) Gaussian(-17,1) ... 0048 Gaussian(-21,1) Gaussian(-19,1) Gaussian(-17,1) Gaussian(-15,1) ... 0049 Gaussian(-21,1) Gaussian(-19,1) Gaussian(-17,1) Gaussian(-15,1) Gaussian(-13,1) ... 0050 Gaussian( 21,1) Gaussian( 19,1) Gaussian( 17,1) ... 0051 Gaussian( 21,1) Gaussian( 19,1) Gaussian( 17,1) Gaussian( 15,1) ... 0052 Gaussian( 21,1) Gaussian( 19,1) Gaussian( 17,1) Gaussian( 15,1) Gaussian( 13,1)}; 0053 0054 gA={ Gaussian(0,0.05) Gaussian(0,0.05) Gaussian(0,0.05) ... 0055 Gaussian(-2,1) Gaussian(-2,1) Gaussian(-2,1) ... 0056 Gaussian(-4,1) Gaussian(-4,1) Gaussian(-4,1) Gaussian(-4,1)... 0057 Gaussian(-6,1) Gaussian(-6,1) Gaussian(-6,1) Gaussian(-6,1) Gaussian(-6,1)... 0058 Gaussian( 2,1) Gaussian( 2,1) Gaussian( 2,1) ... 0059 Gaussian( 4,1) Gaussian( 4,1) Gaussian( 4,1) Gaussian( 4,1)... 0060 Gaussian( 6,1) Gaussian( 6,1) Gaussian( 6,1) Gaussian( 6,1) Gaussian( 6,1) }; 0061 0062 RM=CS_CA_RewardModel(S,A,w,gS,gA); 0063 0064 if nargin==0 0065 nSampledActions=4; 0066 else 0067 nSampledActions=varargin{1}; 0068 end 0069 0070 % Define the POMDP 0071 POMDP=CS_DO_CA_POMDP('Test1ca',S,A,O,AM,OM,RM,nSampledActions,gamma,ncAlpha); 0072