Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

build videos? #5

Open
nixoye opened this issue Apr 20, 2016 · 28 comments
Open

build videos? #5

nixoye opened this issue Apr 20, 2016 · 28 comments

Comments

@nixoye
Copy link

nixoye commented Apr 20, 2016

are their any plans on making videos on the process of putting one together?

@rockybulwinkle
Copy link
Contributor

Not really. This was a senior design project. As far as I am aware, everyone has moved on. Please note that master_rev2 became our main branch. At some point I am going to merge it into master.

@rockybulwinkle
Copy link
Contributor

All that said, the construction of the frame is fairly simple. If you go to master_rev2, there are pictures.

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

can you link it? also, was there any work on adding gps or wifi to it so it could follow a device? so you could have it trail behind you or be able to get it to a saved location?

@rockybulwinkle
Copy link
Contributor

We had planned on a GPS module. IIRC, it even has a spot where you'd solder it to the PCB. But we didn't get to it before we graduated. In the software's current state, it only supports R/C mode.

@rockybulwinkle
Copy link
Contributor

By R/C mode, I mean dual analog sticks that control throttle, yaw, roll, and pitch.

@rockybulwinkle
Copy link
Contributor

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

oh, wow did you guys not see how amazing this project is and just get the grade forget about it? was the beaglebone detachable from the copter parts easily? also is there also open guides for the control or remote or whatever you called the device to drive it?

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

its so beautiful.

@rockybulwinkle
Copy link
Contributor

No, we get how cool this was ;)

Frankly, after we graduated we didn't have much time to continue development, though some of us wanted to, myself included. I may still pick up this project again. It was my baby, ya know?

it wasn't too hard to separate the bone from the frame. IIRC, you unscrewed four screws and carefully remove it from the PCB.

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

did you guys try carrying anything with it? i hope you do pick this up again!

itd be cool to carry a beaglebone setup as a pc with small keyboard and screen attached with it

@rockybulwinkle
Copy link
Contributor

No, we didn't try carrying anything with it. The controls were a little unwieldily as it was. And I say that as an R/C pilot with more than 15 years of experience.

@rockybulwinkle
Copy link
Contributor

Somehow we had the nerve to fly it in an auditorium full of people XD.

@rockybulwinkle
Copy link
Contributor

Be sure to check out the dropbox videos, if you didn't see them already. I'm the guy with the deeper voice and beard. :p

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

hm, well with gps\cameras itd probably be pretty straightforward to get it to drive itself. it might even be able to use something like openCV to follow directly above you

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

yeah i have the dropbox bookmarked as soon as i came accros this project

@rockybulwinkle
Copy link
Contributor

Are you a Rose-Hulman student as well?

@rockybulwinkle
Copy link
Contributor

"also is there also open guides for the control or remote or whatever you called the device to drive it?"

To control it, you connect to a web server hosted on the bone (it acts as an access point). The website uses some joystick API. We tried using both an xbox controller and a proper R/C controller, specifically the DX6.

@rockybulwinkle
Copy link
Contributor

commands are sent to control_alg by writing to a FIFO and it constantly writes the latest image from the camera to a ramfs. So really you could drive it with a program other than the web server if you wish.

@rockybulwinkle
Copy link
Contributor

If you have any other questions, just ask.

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

"Are you a Rose-Hulman student as well?"
nop, i was looking around for a project i thought i had seen online and came across this.

it'd be interesting to see if you could set it up with a cameras so it could move around like teslas/geohots comma.ai project and be able to have it avoid people\buildings on its own.

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

how durable were they when they fell. did it actually break that time it fell and someone said it broke?
https://www.dropbox.com/sh/0pfk43b3zq14rjh/xk6Kl6lccb#/

@rockybulwinkle
Copy link
Contributor

They were pretty fragile. We did move to plywood/bass wood frames, which both helped the weight and improved durability.

@rockybulwinkle
Copy link
Contributor

Initially it was acrylic.

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

how many did you all build? did you ever look aronud for what GPS you could add to the soldering spot?

@rockybulwinkle
Copy link
Contributor

IIRC, we ordered several of our custom PCBs but only populated maybe 2 or 3. We built many, many frames worth of parts and backups.

@rockybulwinkle
Copy link
Contributor

I forget what GPS we designed it for. It was a specific model. I don't think they all share the same footprint, ya know?

@nixoye
Copy link
Author

nixoye commented Apr 21, 2016

were there any notes or something kept where it might be? i wonder how small cameras can get now that could work with it without putting to much weight. maybe one on each of the things that copter line up the copter wings.

@rockybulwinkle
Copy link
Contributor

Yeah, I'm sure there are notes somewhere. They might be on my google drive, i.e. not public though. The camera was an OV7670 which I wrote a PRU program for. It read the FIFO off the camera and stored it in memory shared with the ARM. It worked well enough, but was a pain to get going. It involved a lot of assembly and magic.

IIRC, we went that route with the camera because we wanted to use the single USB port for networking, and didn't want to have a USB hub strapped to the side.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants