0 # initial state
[4] # terminal state
(0,0,'!A',ConstantRewardFunction(1))
(0,1,'A',ConstantRewardFunction(1))
(1,1,'!A',ConstantRewardFunction(1))
(1,2,'A',ConstantRewardFunction(1))
(2,2,'!B',ConstantRewardFunction(1))
(2,3,'B',ConstantRewardFunction(1))
(3,3,'!C',ConstantRewardFunction(1))
(3,4,'C',ConstantRewardFunction(1))
(4,4,'!D',ConstantRewardFunction(1))
(4,0,'D',ConstantRewardFunction(1))