# print expanded
# policy search
x = goal[0]
y = goal[1]
policy[x][y] = '*'
path = []
path.append([x, y])
while ([x, y] != init):
x1 = x - delta[action[x][y]][0]
y1 = y - delta[action[x][y]][1]
policy[x1][y1] = delta_name[action[x][y]]
x = x1
y = y1
path.append([x, y])
# print policy
path.reverse()
# print path
smooth_path = gridworld.smooth_path(path)
gridworld.draw_path(smooth_path)
gridworld.show()
return smooth_path
smooth_path = run_a_star(grid, heuristics, init, goal, cost=1)
for point in smooth_path:
gridworld.draw_shape("circle", point, 8)
gridworld.show()
time.sleep(1)
gridworld.loop()
