Itake part in the FIRST Robotics Competition, where ach year (beginning in January) thousands of teams composed of high school students are given a game and six weeks to scratch-build a robot that can
effectively play it. These robots are generally around 100
pounds in weight, many reaching the maximum allowed
weight of 120 pounds. A great deal of work goes into
designing, manufacturing, and developing the code for all
of the autonomous functions the robot will need to use.
Six weeks, however, is a very small time constraint to
design, build, and program a complete robot, so each
member of the team needs to spend as much time in the
shop (outside of school, of course) as possible to bring it to
fruition. Many students log over 200 hours at the shop, and
then spend extra time practicing the operation of the robot
and, of course, competing.
Unfortunately, not everyone’s mother is able and/or
willing to drive to the school at 3:00 in the morning to pick
up her child. It’s only natural that most sleep at this hour.
I happened to be one of those kids that could not spend all
kinds of hours at the shop, but somehow I had to be able
to be there.
Necessity and a Dash of
“Why the heck not?”
One day, I discovered the Oculus Rift (Figure 1) — the
amazing virtual reality headset that was both inexpensive
and relatively simple to use. While it wasn’t necessary a way
to solve my problem, it was too cool to pass up. So, I
The initial sketches involved a compact processing unit
that was to stream the video and control the head. The
head was a simple two-servo neck with two webcams on
top. The entire thing would be mounted on some
standardized driving base kit, which I would choose later.
Full of hubris, I was confident that such a simple
arrangement would take no more than a month to build. In
February 2014, I had the first prototype finished and
demonstrated.
The greatest reason for this delay was undoubtedly a
logistical one. The Oculus Rift itself was ordered in June and
came in October. In addition, I cycled through two different
single-board processors — both of which took nearly a
month to arrive. While this was a problem, there was plenty
to work on without the Oculus. I separated the project into
Figure 1. The Oculus Rift DK1.
It's 3:00 a.m. already. I'm still looking over the code, making sure it is as bug
free as I can get it without a robot to test it on. Tomorrow, I will have less
than two hours with the robot before I have to go home. If only I could be
there without having to actually be there physically.
SERVO 11.2014 43
FIRST Things First
Aleksandr Savchenkov is a high school student in La Crescenta,
CA. He participates in the FIRST Robotics competition while
maintaining a 4.0+ GPA. In his spare time, he works on his own
robotics projects, plays guitar, sings, skateboards, and
occasionally sleeps.
By Aleksandr Savchenkov
Post comments on this section and find any
associated files and/or downloads at
www.servomagazine.com/index.php/magazine/
article/november2014_Savchenkov.
