Embedded Systems, CSE 475, Spring 2016

Embedded Systems, CSE 475, Spring 2016


This is the embedded
capstone, CSE 475. This year we’re doing
something unusual. I was talking to
Capitol Hill Housing. They think that if you
could detect open parking spaces in different garages,
overall as a society we wouldn’t have to
build as much parking. And ultimately that would
lead to less expensive housing for people. It occurred to me that the
students could really make some good progress on this. So we took the research money
that they were going to give us and decided to make a prize
money for the students. So they’re all trying to solve
this problem, the same problem, in different ways. And then the prize money will
allow some of the students to keep working on
it in the future after the course is over. And we were really excited
about the opportunity to partner with UW, having a
bunch of students all coming up with
their own ideas and working on
different solutions. And we wanted this
to be something that could become publicly
available for other folks, that was open source. So our group’s the
Parking Ninjas. We’re using magnetometers
to detect cars entering and exiting
the garage to maintain a tally of how many
open spots there are in a particular parking garage. So someone using this
app could look on the app to see where parking
spots are open and then go directly to
that particular garage and park there and
save the time of having a look around for an open spot. We have tested our
system for small metals. We have successfully
collected some data. But we are not sure if we’ll
get the same values for cars, which is our ultimate goal. So today we’re focusing
to find that value to see what’s the
maximum that will not trigger the sensors if it’s
something other than a car. We’re here today to
test our designs. And so how we do this
is by using RFID tags, and they’re like
little stickers. And we stick them
along the bumper. And as cars drive in,
we’re able to detect it. But this thing is something new. It’s like real-high tech techie. The WISP thing, it doesn’t
come consume the power. That’s crazy. So this space can be accessed
from anywhere in the world. So this one is real time now. Currently, we have
two spots available. Very steady at 96% accuracy,
so I think that’s pretty good. The name of our
project is Park N’ Go! And we’re detecting individual
cars in parking spaces to see if they are there or not. And then we’re uploading
that data to the internet. I think the best part about
this project is my team. I think everybody divides
their work equally. And everybody is specialized
in the specific field. Somebody likes software. Somebody likes hardware. We have each other’s back
and help our teammates. I can see the parking data. About where the
parking spaces are available on my actual phone. So we have the camera there
to detect these shapes. There’s a really
small chance there might be another blue
triangle somewhere. So we used two colors. So once a car drives
in, PC can tell. So now a parking
space one says no. It said yes before. And you can see the
car parked over it. That is cool. Well done, wow.

Leave a Reply

Your email address will not be published. Required fields are marked *