1 # This is a Dec-POMDP (.dpomdp) file for the Dec-Tiger problem.
2 # For more detailed documentation, see example.dpomdp
4 # Allright, here we go!
8 #Either 1) the number of agents:
10 #or 2) a list of agent identifiers, e.g.:
11 # agents: agent1_name, name-of-agent2, ...
16 # values: [ reward, cost ]
18 # states: [ %d, <list of states> ]
19 states: tiger-left tiger-right
21 #Examples of this are:
22 # start: 0.3 0.1 0.0 0.2 0.5
26 # start include: first-state third state
28 # start exclude: fifth-state seventh-state
32 #The actions declarations
33 #------------------------
34 #the (number/list of) actions for each of the agents on a separate line
36 # [ %d, <list of actions> ]
37 # [ %d, <list of actions> ]
39 # [ %d, <list of actions> ]
41 listen open-left open-right
42 listen open-left open-right
43 #the (number/list of) observations for each of the agents on a separate line
45 # [ %d, <list of observations> ]
46 # [ %d, <list of observations> ]
48 # [ %d, <list of observations> ]
52 #Transition probabilities
53 # T: <a1 a2...an> : <start-state> : <end-state> : %f
55 # T: <a1 a2...an> : <start-state> :
56 # %f %f ... %f P(s_1'|ja,s) ... P(s_k'|ja,s)
58 # T: <a1 a2...an> : this is a |S| x |S| matrix
59 # %f %f ... %f P(s_1'|ja,s_1) ... P(s_k'|ja,s_1)
62 # %f %f ... %f P(s_1'|ja,s_k) ... P(s_k'|ja,s_k)
65 # [ identity, uniform ]
68 #T:open-right open-right :
72 #Observation probabilities
73 # O: <a1 a2...an> : <end-state> : <o1 o2 ... om> : %f
75 # O: <a1 a2...an> : <end-state> :
76 # %f %f ... %f P(jo_1|ja,s') ... P(jo_x|ja,s')
78 # O:<a1 a2...an> : - a |S|x|JO| matrix
79 # %f %f ... %f P(jo_1|ja,s_1') ... P(jo_x|ja,s_1')
82 # %f %f ... %f P(jo_1|ja,s_k') ... P(jo_x|ja,s_k')
85 O: listen listen : tiger-left : hear-left hear-left : 0.7225
86 O: listen listen : tiger-left : hear-left hear-right : 0.1275
87 O: listen listen : tiger-left : hear-right hear-left : 0.1275
88 O: listen listen : tiger-left : hear-right hear-right : 0.0225
89 O: listen listen : tiger-right : hear-right hear-right : 0.7225
90 O: listen listen : tiger-right : hear-left hear-right : 0.1275
91 O: listen listen : tiger-right : hear-right hear-left : 0.1275
92 O: listen listen : tiger-right : hear-left hear-left : 0.0225
95 # R: <a1 a2...an> : <start-state> : <end-state> :
98 # R: <a1 a2...an> : <start-state> :
104 #Typical problems only use R(s,ja) which is specified by:
105 # R: <a1 a2...an> : <start-state> : * : * : %f
106 R: listen listen: * : * : * : -2
107 R: open-left open-left : tiger-left : * : * : -50
108 R: open-right open-right : tiger-right : * : * : -50
109 R: open-left open-left : tiger-right : * : * : +20
110 R: open-right open-right : tiger-left : * : * : 20
111 R: open-left open-right: tiger-left : * : * : -100
112 R: open-left open-right: tiger-right : * : * : -100
113 R: open-right open-left: tiger-left : * : * : -100
114 R: open-right open-left: tiger-right : * : * : -100
115 R: open-left listen: tiger-left : * : * : -101
116 R: listen open-right: tiger-right : * : * : -101
117 R: listen open-left: tiger-left : * : * : -101
118 R: open-right listen: tiger-right : * : * : -101
119 R: listen open-right: tiger-left : * : * : 9
120 R: listen open-left: tiger-right : * : * : 9
121 R: open-right listen: tiger-left : * : * : 9
122 R: open-left listen: tiger-right : * : * : 9