Playing prince of persia 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 depicts a
human robot playing prince of persia. He is playing
prince of persia for the first time, and has no idea
what the game is about. This video is silent because
I want viewers to pay attention to the robot's
thinking while playing the game.
At the beginning, the
robot reads the instruction manual on the game. The
intelligent pathways in his brain formed a computer
program to populate primarily 4 containers: the task
container, rules container, planning container, and
identity container. Knowledge about videogames and
especially maze games pour into the robot's
conscious. At this point, the tasks in the task
container are known, such as the primary task is to
beat the game and the first task is to finish level
1. In the rules container, common rules on maze games
pour into the container -- things like: prince of
persia is a maze game and the objective is to go from
the current location to a destination location, or
the game has time limits to pass each level, or in
order to open a door you need a key.
Knowledge pours into
the robot's conscious based on 2 things 1.
instructions from the game manual. 2. common
knowledge about the game's genre. The rules and
objectives of this specific game should be known, as
well as, common sense knowledge about this game in
general. As the robot plays the game, data in the 4
containers change. For example, the task container
might have 4 hierarchical tasks instead of 3 or the
rules container might have rules on the 3rd baseman
instead of the pitcher.
In the video, the
robot plays the game 5 times. At first the robot dies
a few times trying to pass level1, but by the time
the robot has played the game 4 times, he knows
"everything" to beat level1. Notice that in the final
scene (the final gameplay) the robot plays the game
flawlessly and beats level 1. He knows exactly what
to do at specific times throughout the final gameplay.
This video shows that
this human robot uses trial and error to play and
beat unknown videogames. This is exactly how human
beings play and beat unknown videogames.
The robot uses a
universal pathway to play the game. What this means
is that this robot can not only play this game, but
any maze game.