Playing resident evil 4 using Human Level
Artificial Intelligence
Note: To make this website free to the public
please click on an ad to support my sponsors or you
can make a tax-deductable donation using Paypal
(click on the donation icon on the left).
This video shows a robot
playing resident evil 4. There are no sound in the
video because I wanted to show the viewers what the
robot is thinking while playing the game. The
flashing text and freeze frames are the internal
thoughts of the robot and not instruction text for
the viewers. These internal thoughts describe the
details of how the robot produce intelligence.
My robot doesn't use:
planning programs/heuristic searches (used by MIT and
Stanford University), Bayesian's probability theories
for decision making, Bayesian's equation for
induction and deduction, semantic networks for
natural language understanding, predicate calculus,
common sense systems, first-order logic, rule-based
systems, genetic programming, or MACHINE LEARNING.
This video shows a robot
playing a complex 3-d game. The whole point of the
video was to show viewers that this robot can
navigate in a 3-d environment based on goals. The
player should be aware of not just the current
viewing state, but the environment from 360 degrees.
The robot has to know where his teammate is at all
times, even when he doesn't see her. The robot has
to know what the street paths looks like so he can
navigate in them. He needs to also track the zombies
and where they are located at all times. This 360
degree awareness of the environment is crucial to
making decisions in the game. For example, in one of
the scenes, he is barricaded inside a house and
zombies are surrounding the house. The robot needs
to know all the exit points and shoot zombies so they
can't enter the house. Some exit points are behind
the robot or to the side. The robot is fully aware
of what is behind him, and to the side (right or
left). This awareness can tell the robot where to
focus his attention and prevent zombies from entering
the house.
The awareness of the
environment is just the first part. The robot is
actually doing multiple tasks at the same time. At
the beginning of the game, the robot has devised a
strategy to stay live for 15 minutes, waiting for
backup. His strategy is to shoot, run, shoot, run,
etc. He knows that if he doesn't kill zombies, he
will be surrounded. Thus, he as no choice, but to
kill zombies. But he has to do it in a way that will
lead him to survival for himself and his teammate.
The shoot and run method works because the player
does survive at the end of the video.
If the robot had decided
to run only and hide, the zombies will come in
streams and eventually find him. If he stands his
ground and shoot zombies at one area, he will be
overwhelmed and killed. Thus, the strategy to run
and shoot, run and shoot, is a method that leads to
survival. This was a strategy the robot discovered
through practice. Through trial and error, he found
the method that worked in the game.