Skip to content

Best-First Search Algorithm in Python

This tutorial shows you how to implement a best-first search algorithm in Python for a grid and a graph. Best-first search is an informed search algorithm as it uses an heuristic to guide the search, it uses an estimation of the cost to the goal as the heuristic.

Best-first search starts in an initial start node and updates neighbor nodes with an estimation of the cost to the goal node, it selects the neighbor with the lowest cost and continues to expand nodes until it reaches the goal node. Best-first search favors nodes that are close to the goal node, this can be implemented by using a priority queue or by sorting the list of open nodes in ascending order. The heuristic should not overestimate the cost to the goal, a heuristic that is closer to the actual cost is better as long as it doesn’t overestimate the cost to the goal.

Best-first search is complete and it will find the shortest path to the goal. A good heuristic can make the search fast, but it may take a long time and consume a lot of memory in a large search space. The time complexity is O(n) in a grid and O(b^d) in a graph/tree with a branching factor (b) and a depth (d). The branching factor is the average number of neighbor nodes that can be expanded from each node and the depth is the average number of levels in a graph/tree.

Grid problem (maze)

I have created a simple maze (download it) with walls, a start (@) and a goal ($). Best-first search is used to find the shortest path from the start node to the goal node by using the distance to the goal node as a heuristic. The distance to the goal node is calculated as the manhattan distance from a node to the goal node.

# This class represents a node
class Node:

    # Initialize the class
    def __init__(self, position:(), parent:()):
        self.position = position
        self.parent = parent
        self.g = 0 # Distance to start node
        self.h = 0 # Distance to goal node
        self.f = 0 # Total cost

    # Compare nodes
    def __eq__(self, other):
        return self.position == other.position

    # Sort nodes
    def __lt__(self, other):
         return self.f < other.f

    # Print node
    def __repr__(self):
        return ('({0},{1})'.format(self.position, self.f))

# Draw a grid
def draw_grid(map, width, height, spacing=2, **kwargs):
    for y in range(height):
        for x in range(width):
            print('%%-%ds' % spacing % draw_tile(map, (x, y), kwargs), end='')
        print()

# Draw a tile
def draw_tile(map, position, kwargs):
    
    # Get the map value
    value = map.get(position)

    # Check if we should print the path
    if 'path' in kwargs and position in kwargs['path']: value = '+'

    # Check if we should print start point
    if 'start' in kwargs and position == kwargs['start']: value = '@'

    # Check if we should print the goal point
    if 'goal' in kwargs and position == kwargs['goal']: value = '$'

    # Return a tile value
    return value 

# Best-first search
def best_first_search(map, start, end):
    
    # Create lists for open nodes and closed nodes
    open = []
    closed = []

    # Create a start node and an goal node
    start_node = Node(start, None)
    goal_node = Node(end, None)

    # Add the start node
    open.append(start_node)
    
    # Loop until the open list is empty
    while len(open) > 0:

        # Sort the open list to get the node with the lowest cost first
        open.sort()

        # Get the node with the lowest cost
        current_node = open.pop(0)

        # Add the current node to the closed list
        closed.append(current_node)
        
        # Check if we have reached the goal, return the path
        if current_node == goal_node:
            path = []
            while current_node != start_node:
                path.append(current_node.position)
                current_node = current_node.parent
            #path.append(start) 
            # Return reversed path
            return path[::-1]

        # Unzip the current node position
        (x, y) = current_node.position

        # Get neighbors
        neighbors = [(x-1, y), (x+1, y), (x, y-1), (x, y+1)]

        # Loop neighbors
        for next in neighbors:

            # Get value from map
            map_value = map.get(next)

            # Check if the node is a wall
            if(map_value == '#'):
                continue

            # Create a neighbor node
            neighbor = Node(next, current_node)

            # Check if the neighbor is in the closed list
            if(neighbor in closed):
                continue

            # Generate heuristics (Manhattan distance)
            neighbor.g = abs(neighbor.position[0] - start_node.position[0]) + abs(neighbor.position[1] - start_node.position[1])
            neighbor.h = abs(neighbor.position[0] - goal_node.position[0]) + abs(neighbor.position[1] - goal_node.position[1])
            neighbor.f = neighbor.h

            # Check if neighbor is in open list and if it has a lower f value
            if(add_to_open(open, neighbor) == True):
                # Everything is green, add neighbor to open list
                open.append(neighbor)

    # Return None, no path is found
    return None

# Check if a neighbor should be added to open list
def add_to_open(open, neighbor):
    for node in open:
        if (neighbor == node and neighbor.f >= node.f):
            return False
    return True

# The main entry point for this module
def main():

    # Get a map (grid)
    map = {}
    chars = ['c']
    start = None
    end = None
    width = 0
    height = 0

    # Open a file
    fp = open('data\\maze.in', 'r')
    
    # Loop until there is no more lines
    while len(chars) > 0:

        # Get chars in a line
        chars = [str(i) for i in fp.readline().strip()]

        # Calculate the width
        width = len(chars) if width == 0 else width

        # Add chars to map
        for x in range(len(chars)):
            map[(x, height)] = chars[x]
            if(chars[x] == '@'):
                start = (x, height)
            elif(chars[x] == '$'):
                end = (x, height)
        
        # Increase the height of the map
        if(len(chars) > 0):
            height += 1

    # Close the file pointer
    fp.close()

    # Find the closest path from start(@) to end($)
    path = best_first_search(map, start, end)
    print()
    print(path)
    print()
    draw_grid(map, width, height, spacing=1, path=path, start=start, goal=end)
    print()
    print('Steps to goal: {0}'.format(len(path)))
    print()

# Tell python to run main method
if __name__ == "__main__": main()
#################################################################################
#.#...#....$....#...................#...#.........#.......#.............#.......#
#.#.#.#.###+###.#########.#########.#.#####.#####.#####.#.#.#######.###.#.#####.#
#...#.....#+++#.#.........#.#.....#.#...#...#...#.......#.#.#.......#.#.#.#...#.#
#############+#.#.#########.#.###.#.###.#.###.#.#.#######.###.#######.#.#.#.#.#.#
#+++++++++++#+#...#.#.....#...#...#...#.#.#.#.#...#...#.......#.......#.#.#.#.#.#
#+#########+#+#####.#.#.#.#.###.#####.#.#.#.#.#####.#.#########.###.###.###.#.#.#
#+#........+#+++#...#.#.#.#...#.....#.#.#.#...#.#...#.......#.....#.#...#...#...#
#+#########+#.#+###.#.#.#####.###.#.#.#.#.#.###.#.#########.#####.#.#.###.#####.#
#+#+++++++#+#.#+++#...#.#.....#.#.#.#...#.#.....#.#.....#.#...#...#.......#...#.#
#+#+#####+#+#.###+#####.#.#####.#.#.###.#.#######.###.#.#.###.#.###########.#.#.#
#+++#+++#+#+#...#+++++#.#.......#.#.#...#.....#...#...#.....#.#.#...#...#...#...#
#####+#+#+#+#########+#.#######.#.###.#######.#.###.#########.###.#.#.#.#.#######
#+++++#+++#+#+++++++++#.......#.#...#.#.#.....#.#.....#.......#...#.#.#.#.#.....#
#+#########+#+#########.###.###.###.#.#.#.###.#.#.###.#.#######.###.#.###.#.###.#
#+++#.#+++++#+++#.....#.#.#...#.#.#.....#...#.#.#...#.#...#...#...#.#.#...#...#.#
###+#.#+#####.#+#.#.###.#.###.#.#.#####.###.###.#####.###.#.#.#.###.#.#.#####.#.#
#+++#+++#.....#+#.#.#...#...#.....#...#.#...#...........#.#.#...#...#.......#.#.#
#+###+#########+#.#.#.###.#.#####.#.#.###.###.###########.#.#####.#########.###.#
#+#..+++++++++++#.#.......#.#...#.#.#...#.#...#.#.......#.......#.#...#.....#...#
#+#.#############.#########.#.#.###.###.#.#.###.#.#####.#.#######.#.#.#.#####.#.#
#+#.#+++++++++++#.#.#.#.....#.#.....#...#.#.....#...#.#.#.#.#...#.#.#.#.#.....#.#
#+###+#########+#.#.#.#######.#######.###.#####.###.#.#.#.#.###.#.#.#.#.#####.#.#
#+++++#+++#+++++#...#.........#.....#...#.....#...#...#.#.....#.#...#.#.#.....#.#
#.#####+#+#+#######.###########.#######.#.#######.###.#.###.###.#####.#.#.#####.#
#.....#+#+#+++#...#.#+++++++#.........#.#...#.......#.#.#...#...#.....#.#.#...#.#
#######+#+###+#.###.#+#####+#.#####.###.#.#.#.#######.#.#####.###.#####.#.###.#.#
#+++++++#+#+++#.....#+#...#+#...#.#.....#.#.#.#.#.....#...#...#...#.....#...#.#.#
#+#######+#+#.#####.#+###.#+###.#.#######.#.#.#.#.#######.#.###.#.###.#####.#.#.#
#+#.#+++++#+#.#+++#.#+++#.#+++#...#.#...#.#...#.#.....#.#...#...#...#.......#...#
#+#.#+#####+#.#+#+#####+#.###+###.#.#.#.#.#####.#####.#.#####.#####.#########.###
#+#..+#..+++#.#+#+#+++#+++#.#+#...#...#.#.#...#.....#...#.#...#...#.....#...#.#.#
#+###+###+#.###+#+#+#+###+#.#+#.#######.#.#.#.#####.###.#.#.###.#.#####.###.#.#.#
#+++#+++#+#.#+++#+#+#+++#+#.#+#.#.......#...#.........#.#...#...#.#...#...#.#...#
#.#+###+#+#.#+###+#+###+#+#.#+#.###.###.###########.###.#.###.###.###.###.#.###.#
#.#+++#+#+#.#+++#+++#+++#+#.#+#.....#...#...#.....#.#...#.....#.....#.#...#...#.#
#.###+#+#+#####+#####+#.#+#.#+#######.###.#.#####.#.#.#############.#.#.###.#.#.#
#...#+#+++#+++#+++++#+#.#+#.#+#+++#...#.#.#.......#.#.#...#...#...#...#.#.#.#...#
###.#+#####+#+#####+#+###+#.#+#+#+#.###.#.#########.#.#.#.#.#.#.#.#####.#.#.#####
#...#+++++++#+++++++#+++++..#+++#+++++++@...........#...#...#...#.......#.......#
#################################################################################

Steps to goal: 339

Graph problem

We have created a graph from a map in this problem, actual distances is used in the graph. The goal is to find the shortest path from one city to another city. We are using a Graph class and a Node class in the best-first search algorithm. We are using straight-line distances (air-travel distances) between cities as our heuristic, these distances will never overestimate the actual distances between cities.

# This class represent a graph
class Graph:

    # Initialize the class
    def __init__(self, graph_dict=None, directed=True):
        self.graph_dict = graph_dict or {}
        self.directed = directed
        if not directed:
            self.make_undirected()

    # Create an undirected graph by adding symmetric edges
    def make_undirected(self):
        for a in list(self.graph_dict.keys()):
            for (b, dist) in self.graph_dict[a].items():
                self.graph_dict.setdefault(b, {})[a] = dist

    # Add a link from A and B of given distance, and also add the inverse link if the graph is undirected
    def connect(self, A, B, distance=1):
        self.graph_dict.setdefault(A, {})[B] = distance
        if not self.directed:
            self.graph_dict.setdefault(B, {})[A] = distance

    # Get neighbors or a neighbor
    def get(self, a, b=None):
        links = self.graph_dict.setdefault(a, {})
        if b is None:
            return links
        else:
            return links.get(b)

    # Return a list of nodes in the graph
    def nodes(self):
        s1 = set([k for k in self.graph_dict.keys()])
        s2 = set([k2 for v in self.graph_dict.values() for k2, v2 in v.items()])
        nodes = s1.union(s2)
        return list(nodes)

# This class represent a node
class Node:

    # Initialize the class
    def __init__(self, name:str, parent:str):
        self.name = name
        self.parent = parent
        self.g = 0 # Distance to start node
        self.h = 0 # Distance to goal node
        self.f = 0 # Total cost

    # Compare nodes
    def __eq__(self, other):
        return self.name == other.name

    # Sort nodes
    def __lt__(self, other):
         return self.f < other.f

    # Print node
    def __repr__(self):
        return ('({0},{1})'.format(self.position, self.f))

# Best-first search
def best_first_search(graph, heuristics, start, end):
    
    # Create lists for open nodes and closed nodes
    open = []
    closed = []

    # Create a start node and an goal node
    start_node = Node(start, None)
    goal_node = Node(end, None)

    # Add the start node
    open.append(start_node)
    
    # Loop until the open list is empty
    while len(open) > 0:

        # Sort the open list to get the node with the lowest cost first
        open.sort()

        # Get the node with the lowest cost
        current_node = open.pop(0)

        # Add the current node to the closed list
        closed.append(current_node)
        
        # Check if we have reached the goal, return the path
        if current_node == goal_node:
            path = []
            while current_node != start_node:
                path.append(current_node.name + ': ' + str(current_node.g))
                current_node = current_node.parent
            path.append(start_node.name + ': ' + str(start_node.g))
            # Return reversed path
            return path[::-1]

        # Get neighbours
        neighbors = graph.get(current_node.name)

        # Loop neighbors
        for key, value in neighbors.items():

            # Create a neighbor node
            neighbor = Node(key, current_node)

            # Check if the neighbor is in the closed list
            if(neighbor in closed):
                continue

            # Calculate cost to goal
            neighbor.g = current_node.g + graph.get(current_node.name, neighbor.name)
            neighbor.h = heuristics.get(neighbor.name)
            neighbor.f = neighbor.h

            # Check if neighbor is in open list and if it has a lower f value
            if(add_to_open(open, neighbor) == True):
                # Everything is green, add neighbor to open list
                open.append(neighbor)

    # Return None, no path is found
    return None

# Check if a neighbor should be added to open list
def add_to_open(open, neighbor):
    for node in open:
        if (neighbor == node and neighbor.f >= node.f):
            return False
    return True

# The main entry point for this module
def main():

    # Create a graph
    graph = Graph()

    # Create graph connections (Actual distance)
    graph.connect('Frankfurt', 'Wurzburg', 111)
    graph.connect('Frankfurt', 'Mannheim', 85)
    graph.connect('Wurzburg', 'Nurnberg', 104)
    graph.connect('Wurzburg', 'Stuttgart', 140)
    graph.connect('Wurzburg', 'Ulm', 183)
    graph.connect('Mannheim', 'Nurnberg', 230)
    graph.connect('Mannheim', 'Karlsruhe', 67)
    graph.connect('Karlsruhe', 'Basel', 191)
    graph.connect('Karlsruhe', 'Stuttgart', 64)
    graph.connect('Nurnberg', 'Ulm', 171)
    graph.connect('Nurnberg', 'Munchen', 170)
    graph.connect('Nurnberg', 'Passau', 220)
    graph.connect('Stuttgart', 'Ulm', 107)
    graph.connect('Basel', 'Bern', 91)
    graph.connect('Basel', 'Zurich', 85)
    graph.connect('Bern', 'Zurich', 120)
    graph.connect('Zurich', 'Memmingen', 184)
    graph.connect('Memmingen', 'Ulm', 55)
    graph.connect('Memmingen', 'Munchen', 115)
    graph.connect('Munchen', 'Ulm', 123)
    graph.connect('Munchen', 'Passau', 189)
    graph.connect('Munchen', 'Rosenheim', 59)
    graph.connect('Rosenheim', 'Salzburg', 81)
    graph.connect('Passau', 'Linz', 102)
    graph.connect('Salzburg', 'Linz', 126)

    # Make graph undirected, create symmetric connections
    graph.make_undirected()

    # Create heuristics (straight-line distance, air-travel distance)
    heuristics = {}
    heuristics['Basel'] = 204
    heuristics['Bern'] = 247
    heuristics['Frankfurt'] = 215
    heuristics['Karlsruhe'] = 137
    heuristics['Linz'] = 318
    heuristics['Mannheim'] = 164
    heuristics['Munchen'] = 120
    heuristics['Memmingen'] = 47
    heuristics['Nurnberg'] = 132
    heuristics['Passau'] = 257
    heuristics['Rosenheim'] = 168
    heuristics['Stuttgart'] = 75
    heuristics['Salzburg'] = 236
    heuristics['Wurzburg'] = 153
    heuristics['Zurich'] = 157
    heuristics['Ulm'] = 0

    # Run search algorithm
    path = best_first_search(graph, heuristics, 'Frankfurt', 'Ulm')
    print(path)
    print()

# Tell python to run main method
if __name__ == "__main__": main()
['Frankfurt: 0', 'Wurzburg: 111', 'Ulm: 294']

Tags:

1 thought on “Best-First Search Algorithm in Python”

  1. Pingback: Artificial Intelligence-16: Best-First Search Algorithm - DAOPOSTS

Leave a Reply

Your email address will not be published. Required fields are marked *