Micro Fly and Jump Jet protocol dissected
Written by bonzo | 
| 
| 
Micro Fly protocol dissected
The Jump Jet and Flying Saucer are RC toys sold by Snell Flight which use the same protocol.
The Microfly uses a pulse-width modulation scheme to transmit a variable throttle signal.
Generating PPM signals with a computer is easy enough (arduino IR or usb uirt), or you could have the computer generate
the data in some other form and use it to drive the standard handset via analog links. Snellflight provides also
a FM receiver for the Jump Jet (but never tested), which would allow you to transmit to it using radio instead of infrared.
You can feed a signal for transmission straight into most hobby transmitters via the buddy box socket,
making this method quite convenient.
With its foam body in place, the Jump Jet is intrinsically self-levelling, so it is not difficult to control
it using position feedback alone - you just need to measure where it is in three dimensions.
Without the foam body the aircraft is a quadcopter, and is not intrinsically self-levelling.
It is possible to control using position feedback only, but you can get much tighter control if you
add a means to provide aircraft attitude feedback as well.
The MicroFly Alien is a tiny RC toy with a simple flying principle: spinning!
A video is worth 1000 words so you can see the small demo:
Dim lights Embed Embed this video on your site
And here the Jump Jet:
Dim lights Embed Embed this video on your site
And here the Flying Saucer:
Dim lights Embed Embed this video on your site
The toy is based on a single DC motor whose speed provides the trust control. The RC control only select the trust and thus control the vertical speed up and down.
The Microfly uses a PWM transmission system. The TX remote command generates pulses at about 200Hz frequency.
The pulse width varies with throttle setting, from 0% (zero throttle) to 25% (max throttle).
The frequency is not critcial, only the percentage duty during which the IR signal is transmitted.
The carrier frequency is 38KHz, but this is also not critical because the receiver acceptance bandwidth is quite wide.
The optical wavelength is 940nm (the most common), and the receiver is filtered for this.
The LED on the remote command is not filtered, the package is a water-clear type.
Feedback control for the MicroFly
I was thinking to use a visual tracking approach to control the height but is actually the hardest task to get right.
You need proportional and differential control terms,
and you need feedback samples at a surprisingly high frequency - 50Hz is often not enough,
which makes the use of video cameras awkward. To take off, the Microfly needs to be given a
short burst of full power to jump it into the air, otherwise it won't rise vertically,
and is best launched from a surface with friction, such as carpet.
Jump Jet and Flying saucer protocol dissected
Feedback control of the Jump Jet
With its foam body in place, the Jump Jet is intrinsically self-levelling, so it is not difficult to control
it using position feedback alone - you just need to measure where it is in three dimensions.
Without the foam body the aircraft is a quadcopter, and is not intrinsically self-levelling.
It is possible to control using position feedback only, but you can get much tighter control if you
add a means to provide aircraft attitude feedback as well.
Last Updated (Wednesday, 07 July 2010 15:14)
