Via IEEE Spectrum, who reports "ETH Zurich has taken that quadcopter fail recovery software that they've been working on that can completely shrug off the loss of an entire propeller, and pulled it out of any sort of controlled or motion capture environment to show that it works in that real world place that we're always going on about. And it's amazing." From the video description:
We have developed an algorithm allowing a quadrocopter to maintain stable flight despite the complete loss of a motor/propeller. In a previous video (http://youtu.be/bsHryqnvyYA) we have shown this algorithm in use, where a motion capture system was used to measure the position and orientation of the quadrocopter.
In this video we have a pilot fly the quadrocopter. The algorithm is executed on the quadrocopter's onboard micro-controller, and the only sensors required are the quadrocopter's angular rate gyroscopes.
We use blinking LEDs, mounted on the quadrocopter's arms, to indicate a virtual yaw angle, so that the pilot can control the vehicle with the same remote control commands after the failure. As an alternative to the LED system, an onboard magnetometer could be used to track the vehicle's yaw angle. Alternatively, using more sophisticated algorithms, the system could be made to work using only the rate gyroscopes.
The failsafe controller uses only hardware that is readily available on a standard quadrocopter, and could thus be implemented as an algorithmic-only upgrade to existing systems. Until now, the only way a multicopter could survive the loss of a propeller or motor is by having redundancy (e.g. hexacopters, octocopters). However, this redundancy comes at the cost of additional structural weight, reducing the vehicle's useful payload. Using this technology, (more efficient) quadrocopters can be used in safety critical applications, because they still have the ability to gracefully recover from a motor/propeller failure.
This technology is patent pending, and we hope that it will soon be available as a product on most commercial platforms.
For more information, please see www.FlyingMachineArena.org.
Researchers:
Mark W. Mueller, Simon Berger, and Raffaello D'Andrea
Pilot:
Thomas Kägi
Video:
Mark W. Mueller, Markus Waibel, and Raffaello D'Andrea
Location:
Flying Machine Arena at ETH Zurich; and Zürichberg, Zürich
Acknowledgements:
This work is supported by and builds upon prior contributions by numerous collaborators in the Flying Machine Arena project. Seehttp://www.flyingmachinearena.org/people.
This work was supported by the SNSF (Swiss National Science Foundation).
Comments
looking at it more carefully it would seem to be a solution based on quaternion math..I suggest you wikipedia same to get more detail as it makes my head hurt to think about this :)
HZL
Wow! Awesome!
Wow! Awesome!! Now we just need a fpv camera (or device in series between camera and video transmitter) that can calculate spinning rate and adjust frame rate of the camera so it snaps a frame at each revolution and provides a (semi) stable "forward" image :)
Hi Chris,
So are you talking to them about licensing this code for APM ? Chris?(I suspect even if you dont answer that you are already, if not you should be!) I would love to see equations or code which describes this in gory detail :)
extremely interesting
HZL
ps this + self tightening or quick lock props(http://www.foxtechfpv.com/propeller-quick-detach-cwccw-p-1303.html) would make quads safe enough to insure(insurance underwriting will govern civic agency and LEO usage i feel) for use by municipal agencies.
This feature in APM would be great
Wow, I was happy when their first demonstration just brought it down safely.
Very interesting progress!
One idea:
Would it be possible to have some kind of free fall decelerator on APM as well as a temporary solution before advanced technology we have seen on this video comes available?
When copter loses a motor and copter begins to fall/spin around working axis, APM needs to detect which motor power is missing. Then it needs to adapt to current situation and try to stay an upper sector as long as possible and when it flips upside down (lower sector) try to climb upper sector again. Every second on upper sector would reduce speed. In fact, APM seems to do almost like this already but it tries to maintain perfect stability and therefore overreacts without knowing motor loss:
Result will be very hard crash but at lower speed than without this function. Reducing speed from for instance 150 mp/h to 90 mp/h would make a difference.
I'm volunteer to test with parachute test copter if someone wants to try it.
Not sure how this technology will end up in APM in the near future as the article indicates they intend on obtaining a patent.
That is amazing and I hope in the future something similar can be incorporated into the APM code! However a propeller, if attached correctly, should not come loose.
Amazing work.