# This is a Dec-POMDP (.dpomdp) file for the broadcast channel problem.

# Allright, here we go!
#
#The agents. 
#----------
#Either 1) the number of agents: 
#   agents: %d
#or 2) a list of agent identifiers, e.g.:
#   agents: agent1_name, name-of-agent2, ... 
agents: 2 
#   discount: %f 
discount: 1 
#.0
#   values: [ reward, cost ] 
values: reward
#   states: [ %d, <list of states> ] 
#   format is Sab where a=1 <-> agent 1 has a message
states: S00 S01 S10 S11
#
#Examples of this are:
#   start: 0.3 0.1 0.0 0.2 0.5
#   start: first-state
#   start: 5
#   start: uniform
#   start include: first-state third state
#   start include: 1 3
#   start exclude: fifth-state seventh-state
#   TODO:check the start-state
start: S11
#
#The actions declarations
#------------------------
#the  (number/list of) actions for each of the agents on a separate line
#   actions: 
#   [ %d, <list of actions> ] 
#   [ %d, <list of actions> ] 
#   ...
#   [ %d, <list of actions> ] 
actions: 
send wait
send wait
#the (number/list of) observations for each of the agents on a separate line
#   observations: 
#   [ %d, <list of observations> ]
#   [ %d, <list of observations> ]
#   ...
#   [ %d, <list of observations> ]
observations: 
Collision No-Collision
Collision No-Collision
#Transition probabilities
#   T: <a1 a2...an> : <start-state> : <end-state> : %f
#or
#   T: <a1 a2...an> : <start-state> :
#   %f %f ... %f			    P(s_1'|ja,s) ... P(s_k'|ja,s)
#or
#   T: <a1 a2...an> :			    this is a |S| x |S| matrix
#   %f %f ... %f			    P(s_1'|ja,s_1) ... P(s_k'|ja,s_1)
#   %f %f ... %f			    ...
#   ...					    ...
#   %f %f ... %f			    P(s_1'|ja,s_k) ... P(s_k'|ja,s_k)
#or
#   T: <a1 a2...an> 
#   [ identity, uniform ]
#
#   The from state does not affect the probabilities of the next state
#   (if action send is performed)
T: send send : * : S00 : 0.09 
#=0.1 * 0.9
T: send send : * : S10 : 0.81 
#=0.9 * 0.9
T: send send : * : S01 : 0.01 
#0.1 * 0.1
T: send send : * : S11 : 0.09 
#0.9 * 0.1
#
T: send wait : S11 : S11 : 0.9 
#0.9 * 1.0
T: send wait : S11 : S01 : 0.1 
#0.1 * 1.0
T: send wait : S01 : S11 : 0.9 
#0.9 * 1.0
T: send wait : S01 : S01 : 0.1 
#0.1 * 1.0
#
T: send wait : S10 : S11 : 0.09  
#0.9 * 0.1
T: send wait : S10 : S01 : 0.01  
#0.1 * 0.1
T: send wait : S00 : S11 : 0.09 
#0.9 * 0.1
T: send wait : S00 : S01 : 0.01 
#0.1 * 0.1
T: send wait : S10 : S10 : 0.81 
#0.9 * 0.9
T: send wait : S10 : S00 : 0.09 
#0.1 * 0.9
T: send wait : S00 : S10 : 0.81 
#0.9 * 0.9
T: send wait : S00 : S00 : 0.09 
#0.1 * 0.9
#
T: wait send : S11 : S11 : 0.1 
#1.0 * 0.1
T: wait send : S11 : S10 : 0.9 
#1.0 * 0.9
T: wait send : S10 : S11 : 0.1 
#1.0 * 0.1
T: wait send : S10 : S10 : 0.9 
#1.0 * 0.9
#
T: wait send : S01 : S11 : 0.09 
#0.9 * 0.1
T: wait send : S01 : S10 : 0.81 
#0.9 * 0.9
T: wait send : S00 : S11 : 0.09 
#0.9 * 0.1
T: wait send : S00 : S10 : 0.81 
#0.9 * 0.9
T: wait send : S01 : S01 : 0.01 
#0.1 * 0.1
T: wait send : S01 : S00 : 0.09 
#0.1 * 0.9
T: wait send : S00 : S01 : 0.01 
#0.1 * 0.1
T: wait send : S00 : S00 : 0.09 
#0.1 * 0.9
#
T: wait wait : S11 : S11 : 1.0 
#1.0 * 1.0
T: wait wait : S01 : S11 : 0.9 
#0.9 * 1.0
T: wait wait : S01 : S01 : 0.1 
#0.1 * 1.0
T: wait wait : S10 : S11 : 0.1 
#1.0 * 0.1
T: wait wait : S10 : S10 : 0.9 
#1.0 * 0.9
T: wait wait : S00 : S11 : 0.09 
#0.9 * 0.1
T: wait wait : S00 : S01 : 0.01 
#0.1 * 0.1
T: wait wait : S00 : S10 : 0.81 
#0.9 * 0.9
T: wait wait : S00 : S00 : 0.09 
#0.1 * 0.9
#
#Observation probabilities
#    O: <a1 a2...an> : <end-state> : <o1 o2 ... om> : %f
#or
#    O: <a1 a2...an> : <end-state> :
#    %f %f ... %f	    P(jo_1|ja,s') ... P(jo_x|ja,s')
#or
#    O:<a1 a2...an> :	    - a |S|x|JO| matrix
#    %f %f ... %f	    P(jo_1|ja,s_1') ... P(jo_x|ja,s_1') 
#    %f %f ... %f	    ... 
#    ...		    ...
#    %f %f ... %f	    P(jo_1|ja,s_k') ... P(jo_x|ja,s_k') 
#
#    what is the factor for the noisy observations?
#    I'm now assuming that P(Collision | send send) = 0.9 per agent
O: * : * : Collision Collision : 0.01 
#
O: * : * : No-Collision Collision : 0.09 
#
O: * : * : Collision No-Collision : 0.09 
#
O: * : * : No-Collision No-Collision : 0.81 
#
O: send send : * : Collision Collision : 0.81 
#
O: send send : * : No-Collision Collision : 0.09 
#
O: send send : * : Collision No-Collision : 0.09 
#
O: send send : * : No-Collision No-Collision : 0.01 
#
#The rewards
#    R: <a1 a2...an> : <start-state> : <end-state> : <o1 o2 ... om> : %f
#or
#or
#    R: <a1 a2...an> : <start-state> : <end-state> :
#    %f %f ... %f
#or
#    R: <a1 a2...an> : <start-state> :
#    %f %f ... %f 
#    %f %f ... %f 
#    ...
#    %f %f ... %f
#
#Typical problems only use R(s,ja) which is specified by:
#   R: <a1 a2...an> : <start-state> : * : * : %f
#
R: send send : * : * : * : 0
R: wait wait : * : * : * : 0
R: send wait : S11 : * : * : 1
R: send wait : S10 : * : * : 1
R: send wait : S01 : * : * : 0
R: send wait : S00 : * : * : 0
R: wait send : S11 : * : * : 1
R: wait send : S10 : * : * : 0
R: wait send : S01 : * : * : 1
R: wait send : S00 : * : * : 0
#