One of my lecturers showed me a model airplane he build from scratch using nothing but styrofoam, duct tape, hot glue, a bit of wood and some R/C stuff.
I also wanted to build something like that but place also a bit more of electronics inside.
So I decided to go for an F-14 ;) The plans on how to cut those planes can be found online.
The first step was to read in the PWM signals from the R/C receiver which turned out to be more difficult than expected as they overlap. I implemented an delta mix for the Elevons. I used 2x1-Axis accelerometers to implement some auto stabilization as I had never have flown an model airplane before ;) I had a lot of problems with vibrations which appeared as massive noise at the output of the accelerometers but I was able to filter that out using a Lowpass filter.
This project is the first step towards a bit more complex UAV and the intention was to gather knowledge about how to build up a (very unstable) plane and how to fly it.
Is there any gain scheduling for the P controller? If not, it's really worth looking into. The control surfaces require smaller inputs are higher airspeeds, so making the gain inversely proportional to airspeed or 3D speed if you don't have a pitot ( speed_3d = sqrt(ground_speed^2 + vert_speed^2) ) really helps with stability. I've only just implemented this into my flight controller and it makes a HUGE difference.
No there is no such thing, just P. The point is that I do not measure airspeed so this is just a small help as I am a total noob in RC flying. For me it was kind of a proof of concept if I make it fly or not :D
Awesome! The F-14 was designed just about a mile from where I live, and I work with many former Grummanites - so that plane is special for me. (It doesn't hurt that it's a great plane too!) Can you upload or link your plans and electronics information? This looks like it would be a great project to build!
hi!
check http://www.rcgroups.com/forums/showthread.php?t=769451 for the plans.
About the electronics. For me it was a challenge to build my own deltamix etc. If you are new to this I would suggest just to use an ordinary R/C system which includes this functionality. If you want also to do it on your own then let me know, I can give you the source codes (don't except high quality code :p)
stabilization was quite easy in a first approach. I simply implemented a P-controller which was taking desired position (from R/C) minus actual position, multiply by some factor and give it to the delta mix. You can see it here: https://www.youtube.com/watch?v=p8IlIoOpW4Q
The next step would be to measure the behaviour and come up with a PID controller but as I have crashed the plane a few month ago I'm currently too lazy to try that ;)
I think that I will start another project for the UAV story, something bigger. Maybe even a zeppelin or something more static to gain knowledge about radio and UAV stuff. Then in the third iteration I can combine this knowledge and build the real thing ;) But this might need some time.