We’re a group of high school seniors attending Nease High School in Jacksonville, FL, who share a passion for building cool stuff. Besides the copter, we also participate in our school’s Botball team. Our desire to build a copter began when we saw some videos of one being built and flown by a group of students at a German university. After dissecting it for hours and reading about other successful designs, we realized that a copter could be made relatively inexpensively, yet be a ton of fun and a great learning experience, not to mention the “epicness” of probably being the youngest quad inventors on the face of the internet. Our ultimate goal for the copter is to be stable, fully autonomous, and carry a camera for vision-assisted tracking and landing purposes.
Matt: I’m the main programmer and circuit designer of the copter, as well as the financier and final owner of the beast. Besides coding, band and music is another huge hobby of mine. When I was younger before I got in to programming, I loved to play with one of those electronics learning labs. The copter is the first time in a while I’ve had to apply electronics, but I really enjoyed it, and am thinking about pursuing a minor in electrical engineering in addition to a major in CS.
Until I get a bio from my friends I’m just going to put a few sentances about them.
Will: Will is the main force behind the copter’s physical construction. Posessing a wide range of tools and skills, he has been able to creatively engineer solutions to our incredibly skimpy purchases and otherwise lack of any idea of what we’re doing. Without his handyness and incredible garrage full of awesome stuff, not to mention hours upon hours of soldering, this copter simply wouldn’t be possible!
Jeff: Jeff has been a big help simply in following along and thinking things through that extra bit further to realize “Oh yeah, that probably won’t work.” Jeff is going to help construct our pan-tilt camera mount using Botball-style lego construction. Jeff posseses some incredible artistic talent, and has also been able to produce concept sketches and visualize landing gear and the overal frame of the quad.
Karsh: Utkarsh (but we call him Karsh), was interested in the quad from the beginning, and in the course of attending a few meetings learned to solder and split some of the soldering workload with Will. Karsh also does programming on the Botball team, and when I get a chance I think he needs to get a linux install up and learn to hack on the quadcopter code as well! 
Nate: Nate is homeschooled, though a member of our Botball team. He’s a pretty busy dude, but was able to make it to one summer meeting of the copter and plunged head-first in to transistors, schematic design, and a lot of soldering.
Hi Matt,
interesting project you guys are doing. Were you yet aware that the paparazzi guys have a similar project going at http://paparazzi.enac.fr/wiki/Booz with loads of seriously advanced GPL’d tools, like simulators and such? Wicked stuff. Of course designing and building it yourself is as educational as it is fun.
I happen to be going down a similar path and may have a bit of STM32 code here that might just help you guys going a bit more. It’s a command line interface using the USB CDC interface. I Don’t know if you’ve got something like that up and running yet. If you don’t then it will most probably be quite useful to you. I’ll publish it on my site some time soon. Contact me by mail if you’re interested, and I’ll send you a copy.
You can find similar code here, but it doesn’t yet include any STM32 specific code, but integrating it into your own code is real easy:
http://sourceforge.net/projects/tinysh/
Cheers,
Johnny
Hey Johnny,
I think I’d read/seen the paparazzi project somewhere but didn’t know that it was being extended to quads, thanks for the link. Another successful design to study!
I realize I haven’t yet made a solid post talking about the plan behind the software, but at the present we don’t do anything with the USB port because I haven’t really found a reason to use it, and we’re so pressed for pins that we’ll actually using USBDP and USBDM as GPIOs to drive the LCD shift register I think. We do both programming, using the BOOT0 pin high to trigger the stm32′s internal bootloader, and communications via UART1. If we were going to do this again I’d definitely slap an AVR on the I2C bus to handle things like pots, LCDs, landing sensors, etc, all of which I’ve allocated pins for when we get around to implementing them. We’ve actually got some planned use for all of the pins on the header board.
A command line interface would be something pretty cool to have, but I haven’t yet written/added one just because I don’t think our system would really have much use for it. At the present our codebase compiles to a few different test binaries that run do things like run kalman filters or output sensor readings; there’s no “OS” or even complete flight control loop yet. Since flashing takes only a second or so I usually just swap test programs when I want to work on different parts of the code. Is there some advantage to having a command line that I’m missing? I’ve got newlib system calls using UART1 so any ordinary command line C program runs, I just find the current system of multiple test binaries to be working well enough.
Thanks for the tips!
Matt
Hiya,
I use the USB port on my little Olimex board because that way I can connect to any old PC without a UART converter. I like that. A commandline is indeed not necessary, but I find it very nice to be able to modify and monitor parameters of my projects on the fly. All of this can be done with buttons and an LCD, but having a commandline with history and line editing is lots more comfortable and requires zero extra hardware. Ofcourse, it’s a matter of taste.
Cheers,
Johnny
Hi, I am building a quad copter as a degree project, I managed to get the angle from the gyro and accel, I am finding it a bit difficult for the PID part. At the time being, I am using the proportinal part only and it is really slow in reacting. Could you please help me?
Hi Guys,
We are a group of Electrical engineering students at St. Cloud State University at St. Cloud, MN, USA. We are working on an autonomous quadcopter with the ability to hover identical to what you have done. We looked at your website and it is very impressive with what you have accomplished. We need a little suggestion and feedback from you so we can accomplish this project. Currently, we have a 9 DOF IMU which can give us a precise tilt and angle for roll, pitch and yaw. We are currently trying to balance one axis by keeping the other axis fixed on a stand similar to what you have done. We applied a simple PD control with IMU as a feedback to balance in one axis. When we set our reference angle to be 0 degrees and test it in our system, it does not remain at 0 and there is always an offset. Also there is constant oscillation in the steady state angle. We have tried changing the parameter of our P and D values but could not come to perfect balancing. We are new in Control system and we were wondering if you could help us on how to implement the controls system for this project.
Any little help would be appreciated.
Thank you
Guinness