Controlling an excavator 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).
DARPA, a defense agency
of the US government has announced the Robot
Challenge 2012 and the first contest will be held on
Dec. of 2013. Their mission is to give money to
Universities and technology companies to build them a
robot that can do "mundane human tasks". According
to the government, the reason for the the contest was
to build robots that can do search and rescue
missions. They wanted to build robots that can go
into dangerous zones to do tasks. The Japan
earthquake was one example.
There were several tasks
(about 8 tasks) the human robot has to accomplish in
the robot challenge. 1. walk around dangerous
terrains. 2 open a locked door. 3. identify a
broken pipe. 4. climb a ladder. 5. use tools,
like a sledge hammer. 6. fix a broken pipe. 7.
control vehicles with simple functions. 8. push
buttons.
In the video, a robot
was given a task to install gas pipelines on the
entire stretch of H2 highway. The robot has about a
month to dig and install the pipelines (22 miles). I
know that it doesn't require a lot of intelligence to
operate an excavator, and that a 5 year old child can
probably do this. However, there are things that
only adults understand and operating construction
machines has great responsibilities. If a person
uses an excavator on a very heavy load and it exceeds
the weight capacity, then the bucket can break. This
will cost the company thousands of dollars in
repairs. If the propane tank is leaking fluids and
the operator doesn't fix the problem, the excavator
might blow up, killing the driver. Thus, there are
responsibilities that are included in operating an
excavator and only people who are skilled and
certified should operate the machine.
There are no sound in
the video because I wanted to show the viewers what
the robot is thinking while accomplishing a mission.
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.
In the video, the robot
uses lessons in school to operate an excavator.
Controlling an excavator is very easy because there
are only a few controls. However, following standard
procedures, safety inspections, and following rules
while in operation is the most important thing. In
order to operate an excavator, you need to build a
human robot with 2 hands and 2 feet. Those
autonomous excavators in warehouses today are
operated by software (2013). The advantage of using
a human robot is that you can check parts of the
vehicle for damages. The robot can check the tires
to see if there is enough air pressure or check for
damages. If the robot does find damages, he can
replace the tire with his hands and feets. In
addition, a human robot can control any type of
excavator or machine (like a car or plane). You
don't need to buy a specific type of autonomous
excavator. The robot can drive an excavator built in
the 1950's.
Learning to operate an
excavator is done through lectures or books. Thus,
no machine learning is required. If the robot wants
to learn to operate a bulldozer, all he needs is to
read a short instruction manual or by watching other
people operate the machine. Also, there are no
programmers spending 5-10 years writing the internal
codes for each vehicle. The robot has universal
intelligence and can operate any machine, even
machines that are unknown.
Knowledge can also be
shared among different vehicles. For example,
knowledge of driving a car is used to drive an
excavator. Thus, the robot doesn't need to relearn
how to turn the steering wheel or avoid obstacles in
environment. Common sense will allow the robot to
share knowledge for simliar vehicles. For example, a
motorcycle, a truck, and a car are vehicles driven on
the streets. They follow very similar traffic rules
and operate in similar manners. For each vehicle,
the robot will create an optimal computer program in
its mind to operate it. If the robot had to drive a
truck, optimal rules and objectives are activated in
the robot's mind. If the robot had to drive a car,
optimal rules and objectives are activated in the
robot's mind, etc, etc.
By the way, 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. All knowledge in the robot's brain
comes from knowledge taught in school or from books.