def A_star_graph_search(graph: Maze, start, goal):
if goal == start:
return [], []
path = []
path.append(start)
path_heuristic = graph.manhattan_heuristic_calculator(start)
expanded = []
frontier = []
frontier.append(Evaluation(path_heuristic, path))
while True:
item = ExploreFrontier(frontier) # find the best next node to explore
if item is None:
break
current_node = item.getFurthestNode()
current_path = list(item.getPath())
current_cost = item.getCost() - graph.manhattan_heuristic_calculator(
current_node)
expanded.append(current_node)
# if goal was in frontier
if current_node == goal:
return expanded, current_path
nearby_nodes = (graph.getNode(current_node)).adjacent_list()
for node in nearby_nodes:
new_path = list(current_path)
new_path.append(node)
new_cost = current_cost + step_cost + graph.manhattan_heuristic_calculator(
node)
new_item = Evaluation(new_cost, new_path)
location_in_frontier = is_in_frontier(node, frontier)
if location_in_frontier is None:
if node not in expanded:
frontier.append(new_item)
else:
# replace an existing frontier node which have higher cost
if frontier[location_in_frontier].getCost() > new_cost:
frontier.pop(location_in_frontier)
frontier.append(new_item)
# no need to sort frontier 'cause ExploreFrontier can handle it
return None, None
def A_star_graph_search(graph: Maze, start, goal):
if goal == start:
return [goal], [goal]
start_heuristic = graph.manhattan_heuristic_calculator(start)
init = {"Node": start, "Parent": None, "Cost": start_heuristic}
expanded = []
expanded_with_parent = []
frontier = []
frontier.append(init)
while True:
item = ExploreFrontier(frontier)# find the best next node to explore
if item is None:
#print(expanded)
break
expanded.append(item["Node"])
expanded_with_parent.append(item)
# if goal was in frontier
if item["Node"] == goal:
current_path = tracePath(expanded_with_parent, start, item["Node"])
return expanded, current_path
current_cost = item["Cost"] - graph.manhattan_heuristic_calculator(item["Node"])
nearby_nodes = (graph.getNode(item["Node"])).adjacent_list()
for node in nearby_nodes:
new_cost = current_cost + step_cost + graph.manhattan_heuristic_calculator(node)
new_item = {"Node": node, "Parent": item["Node"], "Cost": new_cost}
location_in_frontier = is_in_frontier(node, frontier)
if location_in_frontier is None:
if node not in expanded:
frontier.append(new_item)
else:
temp = frontier[location_in_frontier]
if temp["Cost"] > new_item["Cost"]:
frontier.pop(location_in_frontier)
frontier.append(new_item)
# no need to sort frontier 'cause ExploreFrontier can handle it
return expanded, None