def repeated_backward_a_star(start,goal,grid,Tie_val,backward_expanded_states):
counter = 0
empty_list = list()
while start != goal:
counter += 1
goal.g_val = 0
goal.search_val = counter
start.g_val = infinity
start.search_val = counter
open_list = BHeap(tie_val=Tie_val)
closed_list = list()
goal.f_val = goal.g_val + heuristic(goal,start)
open_list.insert(goal)
optimal_path_reversed, backward_expanded_states = compute_path(goal,start,grid,open_list,closed_list,counter,backward_expanded_states)
optimal_path = list()
for i in range(len(optimal_path_reversed)-1,-1,-1):
item = optimal_path_reversed.pop()
optimal_path.append(item)
if open_list.empty():
print "Cannot reach target"
return empty_list, 0
if not optimal_path:
print "Cannot reach target"
return empty_list, 0
start = optimal_path[0]
for i in range(1,len(optimal_path)):
# if cost(start,optimal_path[i]) != 1:
# # update increased action costs?
# break
# else:
# start = optimal_path[i]
# print "({0},{1})".format(start.x, start.y)
start = optimal_path[i]
print "I reached the target"
return optimal_path, backward_expanded_states
def a_star(start,goal,grid): size = len(grid) closed_list = list() open_list = BHeap() open_list.insert(start) came_from = dict() start.g_val = 0 start.f_val = start.g_val + heuristic(start,goal) while not open_list.empty(): # pdb.set_trace() current = open_list.delete() # get cell with lowest f_value # path.append(current) closed_list.append(current) if current.equals(goal): # pdb.set_trace()def repeated_backward_a_star(start,goal,grid,Tie_val): return reconstruct_path(came_from,goal,size) for neighbor in get_neighbors(current,grid): # pdb.set_trace() if in_closed_list(closed_list,neighbor): # for item in closed_list: # if neighbor.equals(item): continue temp_g_score = current.g_val + 1 # cost(current,neighbor) will always be 1 if temp_g_score < neighbor.g_val: came_from[neighbor.hash_value] = current neighbor.g_val = temp_g_score neighbor.f_val = neighbor.g_val + heuristic(neighbor,goal) if not open_list.contains(neighbor): open_list.insert(neighbor) empty_list = list() return empty_list
def adaptive_a_star(start,goal,grid,Tie_val,adaptive_expanded_states):
counter = 0
empty_list = list()
while start != goal:
counter += 1
start.g_val = 0
start.search_val = counter
goal.g_val = infinity
goal.search_val = counter
open_list = BHeap(tie_val=Tie_val)
closed_list = list()
start.f_val = start.g_val + adaptive_heuristic(start,goal)
open_list.insert(start)
optimal_path, adaptive_expanded_states = compute_adaptive_path(start,goal,grid,open_list,closed_list,counter,adaptive_expanded_states)
if open_list.empty():
print "Cannot reach target"
break
if not optimal_path:
print "Cannot reach target"
break
start = optimal_path[0]
for i in range(1,len(optimal_path)):
# if cost(start,optimal_path[i]) != 1:
# # update increased action costs?
# break
# else:
# start = optimal_path[i]
# print "({0},{1})".format(start.x, start.y)
start = optimal_path[i]
return optimal_path, adaptive_expanded_states
