You could probably argue that most FC´s at the moment is very similar, since we all use the same basic types of HW from the same groups of vendors. How many ways is there to make a wheel?....some do it more elegantly than others and there are differences MCU´s and interfacing, but all is basically doing the same basic stuff with the current popular FC hardwares.
The difference is indeed mostly in the software. AQ is among many other innovations based on a full RTOS with UKF filtering, full featured CAN interfaciong, and adaptive attitude control that is written from the bottom up by Bill Nesbitt and the AQ team.
So I can assure you all that no "copying" is taking place. Anyone in doubt can read the code in our public GLP3 repository. If anything, AQ has supplied a lot of basic research, code and new innovations that is widely beeing taken advantage of out there.
But having said that, I am very sure that AQ M4 is unique in its HW featureset ;-) I dont remember seeing ANY this small controllers with full Nav capabilities...at least none I could buy..;-)
The outline does indeed resemble the Crazyflie. This is intentional, since we build the M4 to be compatible with a number of Nanoquad parts from Hubsan, Crazyflie, Walkera, Wltoys and others. In case of the crazyflie, the outline is a question of allowing clearance between main part of PCB and the rotor face. So its "inspired" from Crazyflie but not a copy...
M4 also fits a Ladybird (and its countless clones) frame with its 20.5mm * 20.5mm M2 mounting pattern.
I also dont beleive any other nanoquad board allows to create hexa or Octo. But with M4, it can be done with a small expansion module:
Pricing details and other relevant info can be found on the AQ forum. At the moment we are gearing up to releasing a very limited number (less than 50) of M4 production protytypes in a public beta. The big release is realisticly still a few months away.
I'm aware about the timeline of OpenPilot / Taulabs,Autoquad a.s.o.
I wrote this only to show up that there is other similar stuff out there.
Years ago I copied Bill's AQ 5 Board for private use... with all the analog stuff this was a lot of interesting work.
Today with all this digital sensor using SPI and I2C it's much easier. Power Supply,CPU, ACC/Gyro, Baro, GPS .. and ready to fly...
In my opinion there remains only one very complex part: the software
I've followed the work of the AutoQuad and OpenPilot Group some years, and I'm sure their results are based on hard work learned by experience and not by copying things (only a short look in the repository shows up, that they use different concepts, and I'm sure crazyflie has their own stuff as well :)
@ThomasB, the similarities are obvious for someone that has worked with the Crazyflie.
Still, i do believe that this will be a nice platform for my research (positionless swarming of drones). @Brian and joebar.rc, any guesses on the price?
"I had a look at the crazyflie CPU and Sensorset. I fear that the OpenPilot Project copied the crazyflie project."
I think you have your timelines reversed. The OpenPilot CC/CC3D existed long before the Crazyflie even started to be discussed (at least openly). I ported the OpenPilot firmware to the CrazyFlie "because" they were so similar, and the CC/CC3D hardware/firmware was so well tested.
In the case of CrazyFlie and OpenPilot, I don't think anyone copied anyone. In this case, the components are clearly different, but you can't deny that the shape is strikingly similar.
I had a look at the crazyflie CPU and Sensorset. I fear that the OpenPilot Project copied the crazyflie project.. and that a lot of years ago! Identical sensors, and same cpu (at least in the first hardware versions, later they moved to the F4 like Bill).
Quatos is an implementation of an adaptive controller replacing a traditional PID attitude controller. The quatos features some extra's that will be described when the M4 is available. The m4 as I understood will be delivered with a quatos license. Quatos needs to have a full profile of the multirotor (inertia moments, weights etc) and the ESC's will be controlled not using rpm, but by closed loop thrust.
Within the AutoQuad project it has been used for a long, long time now (guess around a year or two) and proven to be a worthy replacement for a well tuned PID attitude controller, but with more bells 'n whistles..
As said, documentation needs to be written, well for public release anyway, so in time it will be on the side..
Comments
Now finally the first batch of production boards are in the store at Viacopter there´s a special Xmas offer:
Get more info here: http://forum.autoquad.org/viewtopic.php?f=23&t=3965
M4s and expansion boards can be purchased at: https://viacopter.eu/multirotor-shop/autoquad-m4/autoquad-m4-v2-135...
Best wishes from the AutoQuad Team!
Norbert
Is Quatos in the repo? Has Autoquad an acromode now - or will the ladybird stuck to hovering in level mode after the upgrade?
what a shame: "So its "inspired" from Crazyflie but not a copy..."
You could probably argue that most FC´s at the moment is very similar, since we all use the same basic types of HW from the same groups of vendors. How many ways is there to make a wheel?....some do it more elegantly than others and there are differences MCU´s and interfacing, but all is basically doing the same basic stuff with the current popular FC hardwares.
The difference is indeed mostly in the software. AQ is among many other innovations based on a full RTOS with UKF filtering, full featured CAN interfaciong, and adaptive attitude control that is written from the bottom up by Bill Nesbitt and the AQ team.
So I can assure you all that no "copying" is taking place. Anyone in doubt can read the code in our public GLP3 repository. If anything, AQ has supplied a lot of basic research, code and new innovations that is widely beeing taken advantage of out there.
But having said that, I am very sure that AQ M4 is unique in its HW featureset ;-) I dont remember seeing ANY this small controllers with full Nav capabilities...at least none I could buy..;-)
The outline does indeed resemble the Crazyflie. This is intentional, since we build the M4 to be compatible with a number of Nanoquad parts from Hubsan, Crazyflie, Walkera, Wltoys and others. In case of the crazyflie, the outline is a question of allowing clearance between main part of PCB and the rotor face. So its "inspired" from Crazyflie but not a copy...
M4 also fits a Ladybird (and its countless clones) frame with its 20.5mm * 20.5mm M2 mounting pattern.
I also dont beleive any other nanoquad board allows to create hexa or Octo. But with M4, it can be done with a small expansion module:
https://www.youtube.com/watch?v=VqtgMCuTiNA
https://www.youtube.com/watch?v=cNTMeU5LZGA
If you want more, you can use it to control a fullsized rig as shown here by Steve:
https://www.youtube.com/watch?v=79B1EN5Y_3U&list=UUvyI6dkgSbra4...
Pricing details and other relevant info can be found on the AQ forum. At the moment we are gearing up to releasing a very limited number (less than 50) of M4 production protytypes in a public beta. The big release is realisticly still a few months away.
@Brian+Victor
Don't take my last comment too serious ;)
I'm aware about the timeline of OpenPilot / Taulabs,Autoquad a.s.o.
I wrote this only to show up that there is other similar stuff out there.
Years ago I copied Bill's AQ 5 Board for private use... with all the analog stuff this was a lot of interesting work.
Today with all this digital sensor using SPI and I2C it's much easier. Power Supply,CPU, ACC/Gyro, Baro, GPS .. and ready to fly...
In my opinion there remains only one very complex part: the software
I've followed the work of the AutoQuad and OpenPilot Group some years, and I'm sure their results are based on hard work learned by experience and not by copying things (only a short look in the repository shows up, that they use different concepts, and I'm sure crazyflie has their own stuff as well :)
@ThomasB, the similarities are obvious for someone that has worked with the Crazyflie.
Still, i do believe that this will be a nice platform for my research (positionless swarming of drones). @Brian and joebar.rc, any guesses on the price?
"I had a look at the crazyflie CPU and Sensorset. I fear that the OpenPilot Project copied the crazyflie project."
I think you have your timelines reversed. The OpenPilot CC/CC3D existed long before the Crazyflie even started to be discussed (at least openly). I ported the OpenPilot firmware to the CrazyFlie "because" they were so similar, and the CC/CC3D hardware/firmware was so well tested.
In the case of CrazyFlie and OpenPilot, I don't think anyone copied anyone. In this case, the components are clearly different, but you can't deny that the shape is strikingly similar.
Just wondering who in the copterworld wants to buy a PX4 then? Hey Autoquad don't release that!!
@Victor
I had a look at the crazyflie CPU and Sensorset. I fear that the OpenPilot Project copied the crazyflie project.. and that a lot of years ago! Identical sensors, and same cpu (at least in the first hardware versions, later they moved to the F4 like Bill).
Quatos is an implementation of an adaptive controller replacing a traditional PID attitude controller. The quatos features some extra's that will be described when the M4 is available. The m4 as I understood will be delivered with a quatos license. Quatos needs to have a full profile of the multirotor (inertia moments, weights etc) and the ESC's will be controlled not using rpm, but by closed loop thrust.
Within the AutoQuad project it has been used for a long, long time now (guess around a year or two) and proven to be a worthy replacement for a well tuned PID attitude controller, but with more bells 'n whistles..
As said, documentation needs to be written, well for public release anyway, so in time it will be on the side..
-
1
-
2
-
3
of 3 Next