"""Implement Agents and Environments (Chapters 1-2).
The class hierarchies are as follows:
Thing ## A physical object that can exist in an environment
Agent
Wumpus
Dirt
Wall
...
Environment ## An environment holds objects, runs simulations
XYEnvironment
VacuumEnvironment
WumpusEnvironment
An agent program is a callable instance, taking percepts and choosing actions
SimpleReflexAgentProgram
...
EnvGUI ## A window with a graphical representation of the Environment
EnvToolbar ## contains buttons for controlling EnvGUI
EnvCanvas ## Canvas to display the environment of an EnvGUI
"""
# TO DO:
# Implement grabbing correctly.
# When an object is grabbed, does it still have a location?
# What if it is released?
# What if the grabbed or the grabber is deleted?
# What if the grabber moves?
#
# Speed control in GUI does not have any effect -- fix it.
from grid import distance2
from statistics import mean
import random
import copy
import collections
# ______________________________________________________________________________
class Thing(object):
"""This represents any physical object that can appear in an Environment.
You subclass Thing to get the things you want. Each thing can have a
.__name__ slot (used for output only)."""
def __repr__(self):
return ''.format(getattr(self, '__name__', self.__class__.__name__))
def is_alive(self):
"Things that are 'alive' should return true."
return hasattr(self, 'alive') and self.alive
def show_state(self):
"Display the agent's internal state. Subclasses should override."
print("I don't know how to show_state.")
def display(self, canvas, x, y, width, height):
# Do we need this?
"Display an image of this Thing on the canvas."
pass
class Agent(Thing):
"""An Agent is a subclass of Thing with one required slot,
.program, which should hold a function that takes one argument, the
percept, and returns an action. (What counts as a percept or action
will depend on the specific environment in which the agent exists.)
Note that 'program' is a slot, not a method. If it were a method,
then the program could 'cheat' and look at aspects of the agent.
It's not supposed to do that: the program can only look at the
percepts. An agent program that needs a model of the world (and of
the agent itself) will have to build and maintain its own model.
There is an optional slot, .performance, which is a number giving
the performance measure of the agent in its environment."""
def __init__(self, program=None):
self.alive = True
self.bump = False
self.holding = []
self.performance = 0
if program is None:
def program(percept):
return eval(input('Percept={}; action? ' .format(percept)))
assert isinstance(program, collections.Callable)
self.program = program
def can_grab(self, thing):
"""Returns True if this agent can grab this thing.
Override for appropriate subclasses of Agent and Thing."""
return False
def TraceAgent(agent):
"""Wrap the agent's program to print its input and output. This will let
you see what the agent is doing in the environment."""
old_program = agent.program
def new_program(percept):
action = old_program(percept)
print('{} perceives {} and does {}'.format(agent, percept, action))
return action
agent.program = new_program
return agent
# ______________________________________________________________________________
def TableDrivenAgentProgram(table):
"""This agent selects an action based on the percept sequence.
It is practical only for tiny domains.
To customize it, provide as table a dictionary of all
{percept_sequence:action} pairs. [Figure 2.7]"""
percepts = []
def program(percept):
percepts.append(percept)
action = table.get(tuple(percepts))
return action
return program
def RandomAgentProgram(actions):
"An agent that chooses an action at random, ignoring all percepts."
return lambda percept: random.choice(actions)
# ______________________________________________________________________________
def SimpleReflexAgentProgram(rules, interpret_input):
"This agent takes action based solely on the percept. [Figure 2.10]"
def program(percept):
state = interpret_input(percept)
rule = rule_match(state, rules)
action = rule.action
return action
return program
def ModelBasedReflexAgentProgram(rules, update_state):
"This agent takes action based on the percept and state. [Figure 2.12]"
def program(percept):
program.state = update_state(program.state, program.action, percept)
rule = rule_match(program.state, rules)
action = rule.action
return action
program.state = program.action = None
return program
def rule_match(state, rules):
"Find the first rule that matches state."
for rule in rules:
if rule.matches(state):
return rule
# ______________________________________________________________________________
loc_A, loc_B = (0, 0), (1, 0) # The two locations for the Vacuum world
def RandomVacuumAgent():
"Randomly choose one of the actions from the vacuum environment."
return Agent(RandomAgentProgram(['Right', 'Left', 'Suck', 'NoOp']))
def TableDrivenVacuumAgent():
"[Figure 2.3]"
table = {((loc_A, 'Clean'),): 'Right',
((loc_A, 'Dirty'),): 'Suck',
((loc_B, 'Clean'),): 'Left',
((loc_B, 'Dirty'),): 'Suck',
((loc_A, 'Clean'), (loc_A, 'Clean')): 'Right',
((loc_A, 'Clean'), (loc_A, 'Dirty')): 'Suck',
# ...
((loc_A, 'Clean'), (loc_A, 'Clean'), (loc_A, 'Clean')): 'Right',
((loc_A, 'Clean'), (loc_A, 'Clean'), (loc_A, 'Dirty')): 'Suck',
# ...
}
return Agent(TableDrivenAgentProgram(table))
def ReflexVacuumAgent():
"A reflex agent for the two-state vacuum environment. [Figure 2.8]"
def program(percept):
location, status = percept
if status == 'Dirty':
return 'Suck'
elif location == loc_A:
return 'Right'
elif location == loc_B:
return 'Left'
return Agent(program)
def ModelBasedVacuumAgent():
"An agent that keeps track of what locations are clean or dirty."
model = {loc_A: None, loc_B: None}
def program(percept):
"Same as ReflexVacuumAgent, except if everything is clean, do NoOp."
location, status = percept
model[location] = status # Update the model here
if model[loc_A] == model[loc_B] == 'Clean':
return 'NoOp'
elif status == 'Dirty':
return 'Suck'
elif location == loc_A:
return 'Right'
elif location == loc_B:
return 'Left'
return Agent(program)
# ______________________________________________________________________________
class Environment(object):
"""Abstract class representing an Environment. 'Real' Environment classes
inherit from this. Your Environment will typically need to implement:
percept: Define the percept that an agent sees.
execute_action: Define the effects of executing an action.
Also update the agent.performance slot.
The environment keeps a list of .things and .agents (which is a subset
of .things). Each agent has a .performance slot, initialized to 0.
Each thing has a .location slot, even though some environments may not
need this."""
def __init__(self):
self.things = []
self.agents = []
def thing_classes(self):
return [] # List of classes that can go into environment
def percept(self, agent):
'''
Return the percept that the agent sees at this point.
(Implement this.)
'''
raise NotImplementedError
def execute_action(self, agent, action):
"Change the world to reflect this action. (Implement this.)"
raise NotImplementedError
def default_location(self, thing):
"Default location to place a new thing with unspecified location."
return None
def exogenous_change(self):
"If there is spontaneous change in the world, override this."
pass
def is_done(self):
"By default, we're done when we can't find a live agent."
return not any(agent.is_alive() for agent in self.agents)
def step(self):
"""Run the environment for one time step. If the
actions and exogenous changes are independent, this method will
do. If there are interactions between them, you'll need to
override this method."""
if not self.is_done():
actions = []
for agent in self.agents:
if agent.alive:
actions.append(agent.program(self.percept(agent)))
else:
actions.append("")
for (agent, action) in zip(self.agents, actions):
self.execute_action(agent, action)
self.exogenous_change()
def run(self, steps=1000):
"Run the Environment for given number of time steps."
for step in range(steps):
if self.is_done():
return
self.step()
def list_things_at(self, location, tclass=Thing):
"Return all things exactly at a given location."
return [thing for thing in self.things
if thing.location == location and isinstance(thing, tclass)]
def some_things_at(self, location, tclass=Thing):
"""Return true if at least one of the things at location
is an instance of class tclass (or a subclass)."""
return self.list_things_at(location, tclass) != []
def add_thing(self, thing, location=None):
"""Add a thing to the environment, setting its location. For
convenience, if thing is an agent program we make a new agent
for it. (Shouldn't need to override this."""
if not isinstance(thing, Thing):
thing = Agent(thing)
assert thing not in self.things, "Don't add the same thing twice"
thing.location = location if location is not None else self.default_location(thing)
self.things.append(thing)
if isinstance(thing, Agent):
thing.performance = 0
self.agents.append(thing)
def delete_thing(self, thing):
"""Remove a thing from the environment."""
try:
self.things.remove(thing)
except ValueError as e:
print(e)
print(" in Environment delete_thing")
print(" Thing to be removed: {} at {}" .format(thing, thing.location))
print(" from list: {}" .format([(thing, thing.location) for thing in self.things]))
if thing in self.agents:
self.agents.remove(thing)
class Direction():
'''A direction class for agents that want to move in a 2D plane
Usage:
d = Direction("Down")
To change directions:
d = d + "right" or d = d + Direction.R #Both do the same thing
Note that the argument to __add__ must be a string and not a Direction object.
Also, it (the argument) can only be right or left. '''
R = "right"
L = "left"
U = "up"
D = "down"
def __init__(self, direction):
self.direction = direction
def __add__(self, heading):
if self.direction == self.R:
return{
self.R: Direction(self.D),
self.L: Direction(self.U),
}.get(heading, None)
elif self.direction == self.L:
return{
self.R: Direction(self.U),
self.L: Direction(self.L),
}.get(heading, None)
elif self.direction == self.U:
return{
self.R: Direction(self.R),
self.L: Direction(self.L),
}.get(heading, None)
elif self.direction == self.D:
return{
self.R: Direction(self.L),
self.L: Direction(self.R),
}.get(heading, None)
def move_forward(self, from_location):
x, y = from_location
if self.direction == self.R:
return (x+1, y)
elif self.direction == self.L:
return (x-1, y)
elif self.direction == self.U:
return (x, y-1)
elif self.direction == self.D:
return (x, y+1)
class XYEnvironment(Environment):
"""This class is for environments on a 2D plane, with locations
labelled by (x, y) points, either discrete or continuous.
Agents perceive things within a radius. Each agent in the
environment has a .location slot which should be a location such
as (0, 1), and a .holding slot, which should be a list of things
that are held."""
def __init__(self, width=10, height=10):
super(XYEnvironment, self).__init__()
self.width = width
self.height = height
self.observers = []
# Sets iteration start and end (no walls).
self.x_start, self.y_start = (0, 0)
self.x_end, self.y_end = (self.width, self.height)
perceptible_distance = 1
def things_near(self, location, radius=None):
"Return all things within radius of location."
if radius is None:
radius = self.perceptible_distance
radius2 = radius * radius
return [(thing, radius2 - distance2(location, thing.location)) for thing in self.things
if distance2(location, thing.location)