A new flight video for the CrazyFlie.
This entry was posted
on Thursday, April 21st, 2011 at 11:11 and is filed under Crazy Flie, CrazyCopter.
You can follow any responses to this entry through the RSS 2.0 feed.
Both comments and pings are currently closed.
Very nice little copter !
Does the Cortex-M3 have enough power to do the calculations itself ?
Do you have an idea of the overall cost of this project ?
Very good work on this !
Cute one ) … i realy love that micro-designs
also a nice and vers smale one.
do you plan to sell kits in near future?
I’d love to buy one of these!
Awesome!! If you guys were to ever sell this in a kit, let me know! I’ll be sure to get one!
@Montspy: The Cortex-M3 have more than enough power for a basic PID regulation (we run it @64MHz and we are doing all calculation using floating point numbers). As of beginning 2010 the material cost was of about 70-80€, the gyros are less expensive now (the most expensive is the motors and propellers because we had to buy RC plane to get them, we plan to source them directly in the future)
Awesome stuff! Please let me know if you ever plan on selling these. I love that the PS3 controller can be used for control! Excellent work!
Great work guys! I hope you guys make this open source & post the details. It’s gonna become a platform for a breed of awesome robots. Any plans to diversify in the wireless part? Zigbee? Wifi?
This is just amazing! Amazingly tiny too! I know I’m not the only one with possibilities flying around in my head right now.
Are the motors brushless DC motors?
Kudos guys! There are lots of interested people!
One more thing guys. Here’s a feature that I think would completely revolutionize this little critter; first apps for Android, iOS, WebOS, BB OS & WP7 since nearly every phone has Bluetooth built in nowadays.
Hej! Vill ni har hjälp med att göra en WP7-app till den där?
This is really cool haha. I want one!!
I hope your team keeps it forward momentum, because I’d love to follow your progress!
I would suggest to cut down on the flexibility of the arms WITHOUT increasing the weight at all by just rotating the central square board section 45 degrees. This should help with the platform’s stability, and with its durability.
Obviously the layout would need to be tweaked due to the different power exit points to the skinny motor arms, but it should be reasonable.
Let me know what you think.
Also, if the remaining arms are a bit too flexible, then make several pads of copper exposed on one edge of each arm, and get another SKINNY piece of PWB with matching copper pads and solder it on at right angles to the main board’s arm – giving a lightweight CHEAP stiff right-angle cross-sectioned arm.
Best of luck!
Also, I would suggest staying away from brush motors – the simpler electronics are offset by too much brush maintenance, IMO.
This is so cool, I’m really fascinated !
I wanna buy one NOW
Keep it fun to get more progress…
May I Suggest?
I have a quick pick in data structure. 1st joystick, than Computer and than Copter.
Check it out. I gess u´ll have a more stable flight if u develop a “smoother decelerate”. Will prevent crashes too. Get easy to land.
Tks…. and Goooooood work… realy cool.
Great work! How do the accelerometers handle the motor vibrations?
Great Job.Waiting for some new features …
Thanks for your input Ro! We have been thinking about that kind of board but for our first design we wanted do disturb the air flow as little as possible but with the smallest length. Since it turned out to have plenty of lifting force that kind of design seams like a next good step.
If you have accelerometers and Cortex-M3 why do you still need the Gyro?
I would like to build one myself – please explain.
The accelerometer is subject to the motors vibration and the copter acceleration so it cannot be used to have the instant copter pitch/roll angles. However if we filter the accelerometer (ie. mean it over a couple of seconds) we get a pretty good precision on the pitch/roll angles, but the signal is then too slow to permits to use it for the regulation. The gyros, which almost not sensitive to vibration, are used for instant angle calculation and the mean value of the accelerometer is used to recalibrate the gyro over a longer period (this is called sensor fusion).
The Z gyro for the Yaw is working alone because the accelerometer cannot physically sense rotation around the gravity vector (the yaw). This could actually be a problem because this kind of gyro tends to drift with time (however it is pretty OK over the flight period of 4 minutes).
Daedalus Projects is proudly powered by
WordPress and sponsored by Epsilon.
and Comments (RSS).
Switch to our mobile site