After going over a lot of projects here on DIYDrones I realized that there are not a lot of projects involving thrust vector vehicle control.
Inspired by among others the guys at Armadillo and by following my local heroes at CopenhagenSuborbitals I decided to master the technique of balancing a vehicle on a column of exhaust, wither it being air, rocket blast, water jet or something completely different.
Here is one of my initial experiments the Reactive Control Test Platform:
The RCTP uses 3 vanes, spaced 120 degrees apart, to direct the airflow from a single propeller. The on board IMU (ArduIMU V2) receives input from a RC system (Multiplex EVO 9 with Scherrer UHF) and regulates the thrust vector to maintain a pitch, roll and yaw attitude commanded by the RC transmitter.
Why 3 vanes? ... beacuse it is more of a challange :)
All RC signal reception, inertial processing, data logging and servo output are handled by the the ArduIMU V2 on board processor.
The TCTP MK I and II is a pre-study for a bigger and more advanced vehicle that will simulate an ascending rocket.
More pictures and videos of this and other projects on this blog:
http://dzlsevilgeniuslair.blogspot.com/
I am very interested to hear form people that have any experience with similar aircrafts.
Comments
Firmware now running on ArduIMU V3!
A couple of build pictures from V2
Thanks everybody for all the good links.
It looks like there are many ways Directed Airflow Controlled Vehicles can be successfully made. With the right controller anything seem to fly.
I'm interested in making really unstable crafts fly like they were sitting on a stick.
Does anybody have have any experience with system identification on the basis of step respone data? I would like to design a stand alone program for control loop tuning for model aircraft for people without access to Matlab.
@David, do you have any more details about your code/tuning process that I can persuade you to share? :)
Cool design! (and greetings from Ålborg, Dzl)
A few related links - the kk guys in Korea have some interesting work along these lines:
Single-prop VTOL https://www.youtube.com/watch?v=6sY3L-dow8c
and a DualCopter with counter-rotating motors https://www.youtube.com/watch?v=-MdvXknHYt8
Of course such things can be bought 'off the shelf' as well:
E-Flite UMX Hyper Taxi https://www.youtube.com/watch?v=7Eukx_MNW14
but where's the fun in that? :-)
KapeteinKuk did something like this, using of course a kkboard: "It is a quick hack of wood spatulas, zip ties, tape, CD control horns, hot glue, depron and one 1.5l plastic bottle."
http://eastbay-rc.blogspot.com/2011/11/kapteinkuks-flying-vtol-bott...
@david. How did you tune your loop controller?
I see that your craft has relatively large surfaces compared to its mass. I imagine this gives you a little more time to collect data before things go wrong.
The RCTP is really heavy and have small surfaces. With even slightly wrong loop filter settings it quickly get very unstable. Since i will be using the control software for other, possibly even more unstable vehicles, I would like to find good methods for determining the regulator parameters.
Did you do any measurements or make any calculations before flying yout craft?
1st time 1 of these has worked, from an individual. The main problem people have is getting enough force from the vanes. The key seems to be the large leading edge wings.
Interesting! Reminds me of the Japanese ball drone builds over on rcgroups.com:
http://www.rcgroups.com/forums/showthread.php?t=1454102