Cooking lobster (1/2) 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 cooking Chinese style lobster dish. There are
no sound in the video because i wanted to show the
viewers what the robot is thinking while cooking
meals in a restaurant. The text flashing on the
screen are the internal instructions (sound data) of
the robot and are not verbal instructions to viewers.
There are several
ways the robot learns how to cook meals: 1. reading
books. 2. watching a demonstration. 3. watching
teacher lectures. Despite the different ways of
learning how to cook, the robot's conscious
identifies vital information from these learning
materials and stores information about cooking meals
in an organized way in memory. For each meal, the
ingredients used and also the standard procedures of
preparing that meal are stored in memory in an
organized way.
Practice is also a
very important tool used to become a skilled chef.
The more practice the robot goes through the better
his cooking skills become.
In the video, the
robot starts off by generating a mental list of
things to do. The robot's conscious made 3 steps to
make this dish: 1. prepare ingredients. 2. cutting
the lobster. 3. cook lobster dish. These 3 steps are
general steps to making this dish. While the robot is
doing each step, his mind will generate a detailed
list of things to do during runtime. For example,
when the robot has to do step3: cook lobster dish, he
will generate 3 other detailed steps, which are: a)
prepare sauce b) cook lobster c) thicken sauce. And
for each sub-step there are even more procedures to
follow. The intelligent pathways in the robot's brain
stores the linear steps to making a lobster dish. The
intelligent pathways in the robot's brain also store
data in a hierarchical manner and data goes from
general to specific.
The robot's conscious
activates the most important steps to make a lobster
dish, but the intelligent pathways provide the
details of what the robot must do (frame-by-frame).
These pathways store motor functions as well as 5
sense data. The robot knows exactly how to move its
hands and body to make the lobster dish because the
intelligent pathways store the robot's body movements
and also store the 5 sense data of what will happen
in the future because of its actions.
This robot might
forget ingredients or forget to do certain steps
while making a dish. For example, at the beginning of
the video, I show the robot generating a list of
ingredients needed for this dish. This list is just a
general list and it doesn't include a detailed
account of every ingredient needed. During the actual
cooking, the robot might remember forgotten
ingredients. This same behavior can happen to linear
steps of cooking as well. The robot might forget
certain steps or do steps non-linearly. Basically,
these are called interruptions. The robot has the
ability to solve interruptions as they occur.
However, the more
practice the robot goes through the less
interruptions he will encounter and the more skillful
he becomes at cooking.
The robot also has
common sense while cooking the lobster dish. During
the cutting of the lobster, the robot accidentally
chopped off a piece that landed on the dirty floor.
The robot has determined that the floor is dirty and
if food touches the floor, the food becomes dirty
too. Based on logic, the robot has concluded that the
lobster piece, if ingested by a human being, might
get sick. The robot attempts to solve this problem by
coming up with a solution. He proposes to clean the
lobster piece and to get rid of the dirt, which is
what most chefs would do.
Lastly, this robot
has the ability to cook any meal for any restaurant,
regardless of complexity of meals. He also has the
ability to learn cooking skills through different
ways. He can read books, attend cooking lectures or
view a cooking demonstration on TV. This robot chef
can work in a Chinese restaurant, a Mexican
restaurant, a Hawaiian restaurant or any restaurant.
It's adaptive skills allow it to learn and cook any
food.
In upcoming videos, i
will show how the robot chef can cook and manage
multiple meals at the same time. For example, the
robot can cook a lobster dish, a cabbage dish, and
corn soup at the same time.