0], [0, 1, 0, 0, 0, 0, 0, 0, 1, 0],
[0, 1, 0, 1, 1, 0, 1, 0, 1,
0], [0, 1, 0, 1, 0, 0, 1, 0, 1, 0],
[0, 1, 1, 1, 0, 1, 1, 0, 1,
0], [0, 0, 0, 0, 0, 1, 0, 0, 0, 0],
[0, 0, 1, 0, 1, 1, 0, 0, 1, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 1, 0]]))
# =============================
# usage of showState(p):
p = np.zeros(myMaze.stateSize)
p[0] = 0.2
p[2] = 0.3
p[8] = 0.5
pShow(p, myMaze) #Note that pShow scales the probability so that the
#Maximum values is 1 this makes it helpful to visuzliaze
#probabilities that are thinly spread out
# =============================
rob = robot(myMaze)
rob.prob = p
robotShow(rob)
rob.step()
rob.step()
rob.step()
robotShow(rob)
''' Set rob.prob to the steady state'''
rob.randomize()
0], [0, 1, 0, 0, 0, 0, 0, 0, 1, 0],
[0, 1, 0, 1, 1, 0, 1, 0, 1,
0], [0, 1, 0, 1, 0, 0, 1, 0, 1, 0],
[0, 1, 1, 1, 0, 1, 1, 0, 1,
0], [0, 1, 0, 0, 0, 1, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 1, 0]]))
stateReward = np.array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 100, 0, 0, 0, 0, 0],
[
-1000, -1000, -1000, -1000, -1000, -1000,
-1000, -1000, -1000, -1000
]])
mdp = MDPmaze(myMaze, stateReward)
iterCount = 100
printSkip = 10
for i in range(iterCount):
mdp.valIter()
if np.mod(i, printSkip) == 0:
print("Iteration ", i)
pShow(mdp.value, myMaze)
[0, 1, 0, 1, 1, 0, 1, 0, 1,
0], [0, 1, 0, 1, 0, 0, 1, 0, 1, 0],
[0, 1, 1, 1, 0, 1, 1, 0, 1,
0], [0, 1, 0, 0, 0, 1, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 1, 0]]))
stateReward = np.array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 100, 0, 0, 0, 0, 0],
[
-1000, -1000, -1000, -1000, -1000, -1000,
-1000, -1000, -1000, -1000
]])
mdp = MDPmaze(myMaze, stateReward)
iterCount = 1000
printSkip = 100
v = np.zeros(mdp.stateSize)
for i in range(iterCount):
v = mdp.valIter()
if np.mod(i, printSkip) == 0:
print("Iteration ", i)
pShow(v, myMaze)
[1.6, 3.9, 2.5, 4.9, 3.9, 6.7, 9.4, 1.0, 7.1, 6.8],
[7.7, 9., 9.7, 9.3, 0.4, 6.5, 3.9, 0, 1.7, 2.6],
[-9000, 2.1, 8.8, 3.8, 8.4, 8.7, 7.9, 5.8, 9.7, 9.2]
])
ocean = Ocean(oceanMap, oceanReward)
ship = Ship(ocean)
'''
Show probability of ship sailing north from harbor once (before/after)
'''
#Harbor coordinates
harborState = ocean.coord2state((3, 2))
print("Reward for Harbor: %f" % oceanReward[3, 2])
#set location probability for sihp to harbor state
ship.prob = np.zeros(ocean.stateSize)
ship.prob[harborState] = 1
pShow(ship.prob, ocean) #probability 1 of begin in harbor
ship.sail('N')
pShow(ship.prob, ocean) #probability after sailing north from harbor
mdp = MDPSailing(ship)
#Run Value iteration i.e. call valIter(N=something) until mdp.V
#has converged.
'''
Uncomment the line below to once you are ready to save your answers and
want to create the data.json file to upload.
'''
#saveData(mdp)