What do you want from your Flight Control company??

[Bartman]

In the coming months we're going to see new additions to the flight control market that may or may not advance the state-of-the-art. They also may or may not improve what is an increasingly disappointed mood amongst flight control buyers looking for the "Complete Package". We're far enough along with this that we should be smart enough to tell the hype from reality.

So I ask, what do you want from the company that is selling you your next full-featured flight control system?

Here's my list and I look forward to reading what you guys have to say.

From the Company I expect

• Thorough testing and debugging BEFORE posting items for sale, if it doesn't work, don't sell it!
• Clear statement of source regarding parts (if it's a rebranded item I expect people to admit that, rather than BS me about how unique it is)
• Clear and factual documentation of what is included (Hardware AND features)
• Clear and factual documentation of what can reasonably be expected from the equipment (performance)
• Photos and video documentation of the equipment, installed and operating (no photoshop, no stabilization, no tricks at all)
• Detailed information, including photos, showing the parts and how they are to be installed
• Detailed information reviewing the configuration of individual firmware items to get desired performance
• Responsive customer service department, including phone service for emergencies even if it has to be a pay service
• Prompt replacement of defective parts
• Frequent interactive (not just post-and-runs) participation in online forums
• Don't keep adding features if the most basic stuff doesn't work right!

From the equipment I expect

• High quality manufacturing (little bits shouldn't break off of new or hardly used items!!!)
• High quality manufacturing (Worth listing this one twice)
• The most basic features should actually do what they are supposed to do
• Installing hardware shouldn't cause damage to hardware
• Installing the hardware you just paid a lot of money for shouldn't automatically create a presumption of guilt that you caused the defects!!!
• Mounting of components should be straightforward and easy to follow (companies should map it out and make it easy to follow)
• Dual and independent power inputs (no y-connections to one power connection at the FC or power board)
• Multiple receiver or satellite capability
• Manual and auto-level flight modes available on any given flight
• Altitude sensor that is intelligently designed into the system so it doesn't require foam to make it work properly
• Altitude hold that doesn't constantly resemble a yo-yo or a leaky balloon
• Stable in ground effect
• Stable in descending flight
• Stable in wind up to 35 knots
• intelligent throttle response so a hover is a hover rather than a close approximation of a hover
• Compatible with current generation consumer-grade radio systems
• Self monitoring for control or signal failures so as to expedite descent and landing without flying any farther from point of departure
• Ability to set max flight radius to utilize GPS to prevent controlled fly-away events (pilot error basically)
• It's a multi-rotor for God's sake! If a motor fails, prop fails, ESC fails, etc. the FC should be able to manage it and remain controllable! Why have so many motors if it can't do that!?!
• Dynamic GPS positioning (heli is cool with me flying it while it does the GPS thing)
• GPS position hold, come home, and pre-programmed waypoint navigation with google Earth interactivity
• Ability to train and tweak flight controller on the bench to know what level is
• Two LED outputs with configurable functions in firmware (let's face it, the LED's are an absolutely necessary part of multi-rotor helicopters if you're not flying strictly FPV)
• Low voltage warning that you can actually see AND hear, configurable in firmware
• Option to join parts with solder (I don't love servo or bullet connectors, sorry!)
• I actually am starting to prefer the I2c control scheme but I'd like to also see manufacturers bite the bullet and develop ESC's that work specifically with their FC system. I can't believe that Turnigy ESC's are optimized for the control demands of multi-rotor helicopters.
• In the nice to have category, predictive camera stability/control where the flight control can vote on what the camera mount is told to do based on control inputs combined with gyro/acc readings....everything today is reactive and just a little behind what the heli is actually doing

That's my list, thanks for reading. I'm very interested to hear what all y'all have to say about it or add to it. It's inevitable that names will be named but let's avoid a bash-fest of any one company if we could please.

Thanks!
Bart

 

[Benjamin Rowland]

Bart, it seems you've covered almost all of the bases.

I'm a big fan of my FC, but it'd be nice if it came with a case instead of being a bare board. Plastic or aluminum would be cool. As an example, the case on the new Skookum Robotics SK540 flybarless control looks pretty nice.

 

[DucktileMedia]

I would like the fail-safe/return home to have a duration setting for the delay in coming home and how high it goes before returning.

Have separate gains for rate and travel for the gimbal. not mix them together!

 

[Bowley]

Ditto on the gimbal gains

 

[Aerial Media Pros]

I do not want to see a PRE ORDER DISCOUNT for a product that is not fully developed and in production. It looks like allot of people are jumping on the DJI Pre Order band wagon. I dont think its right to take peoples money up front for something your still trying to design or produce. I believe a few companies/dealers are using this marketing to gather thousands in cash flow, just so they can use the funds for the original product developement. Call me crazy, but I think thats messed up. I want to see a fully tested and working product, before offering any type of Pre Order Deals.

 

[DennyR]

I think the idea behind pre order discount should be to get people to start using it and see what it can do. Not to fund the development. What is slightly worrying about s800 is that at NAB it was still in prototype form. The next day they are selling it. An important part of the camera mount is the way in which you attach it to the model That part was not there. It was lashed on. A slight change in the natural frequency of any components could cause problems.

 

[DennyR]

In the coming months we're going to see new additions to the flight control market that may or may not advance the state-of-the-art. They also may or may not improve what is an increasingly disappointed mood amongst flight control buyers looking for the "Complete Package". We're far enough along with this that we should be smart enough to tell the hype from reality.

So I ask, what do you want from the company that is selling you your next full-featured flight control system?

Here's my list and I look forward to reading what you guys have to say.

From the Company I expect

• Thorough testing and debugging BEFORE posting items for sale, if it doesn't work, don't sell it!
• Clear statement of source regarding parts (if it's a rebranded item I expect people to admit that, rather than BS me about how unique it is)
• Clear and factual documentation of what is included (Hardware AND features)
• Clear and factual documentation of what can reasonably be expected from the equipment (performance)
• Photos and video documentation of the equipment, installed and operating (no photoshop, no stabilization, no tricks at all)
• Detailed information, including photos, showing the parts and how they are to be installed
• Detailed information reviewing the configuration of individual firmware items to get desired performance
• Responsive customer service department, including phone service for emergencies even if it has to be a pay service
• Prompt replacement of defective parts
• Frequent interactive (not just post-and-runs) participation in online forums
• Don't keep adding features if the most basic stuff doesn't work right!

From the equipment I expect

• High quality manufacturing (little bits shouldn't break off of new or hardly used items!!!)
• High quality manufacturing (Worth listing this one twice)
• The most basic features should actually do what they are supposed to do
• Installing hardware shouldn't cause damage to hardware
• Installing the hardware you just paid a lot of money for shouldn't automatically create a presumption of guilt that you caused the defects!!!
• Mounting of components should be straightforward and easy to follow (companies should map it out and make it easy to follow)
• Dual and independent power inputs (no y-connections to one power connection at the FC or power board)
• Multiple receiver or satellite capability
• Manual and auto-level flight modes available on any given flight
• Altitude sensor that is intelligently designed into the system so it doesn't require foam to make it work properly
• Altitude hold that doesn't constantly resemble a yo-yo or a leaky balloon
• Stable in ground effect
• Stable in descending flight
• Stable in wind up to 35 knots
• intelligent throttle response so a hover is a hover rather than a close approximation of a hover
• Compatible with current generation consumer-grade radio systems
• Self monitoring for control or signal failures so as to expedite descent and landing without flying any farther from point of departure
• Ability to set max flight radius to utilize GPS to prevent controlled fly-away events (pilot error basically)
• It's a multi-rotor for God's sake! If a motor fails, prop fails, ESC fails, etc. the FC should be able to manage it and remain controllable! Why have so many motors if it can't do that!?!
• Dynamic GPS positioning (heli is cool with me flying it while it does the GPS thing)
• GPS position hold, come home, and pre-programmed waypoint navigation with google Earth interactivity
• Ability to train and tweak flight controller on the bench to know what level is
• Two LED outputs with configurable functions in firmware (let's face it, the LED's are an absolutely necessary part of multi-rotor helicopters if you're not flying strictly FPV)
• Low voltage warning that you can actually see AND hear, configurable in firmware
• Option to join parts with solder (I don't love servo or bullet connectors, sorry!)
• I actually am starting to prefer the I2c control scheme but I'd like to also see manufacturers bite the bullet and develop ESC's that work specifically with their FC system. I can't believe that Turnigy ESC's are optimized for the control demands of multi-rotor helicopters.
• In the nice to have category, predictive camera stability/control where the flight control can vote on what the camera mount is told to do based on control inputs combined with gyro/acc readings....everything today is reactive and just a little behind what the heli is actually doing

That's my list, thanks for reading. I'm very interested to hear what all y'all have to say about it or add to it. It's inevitable that names will be named but let's avoid a bash-fest of any one company if we could please.

Thanks!
Bart

Flight control as we know it is all going to change. It will become highly configerable Data transfer. Like OSRC/ Ruling. Your receiver will do everything.

 

[jes1111]

Good list!

I'm going to mention a name, but it should be obvious that I have no commercial motive - the OpenPilot Project has a similar (possibly longer) goal list with one or two very significant additions: like "lowest cost possible" and "open source". The (aptly named) OP Revolution was flying (in production hardware form) at the recent developers' meet in Portugal - think "CopterControl on steroids with full navigation and autonomy features". I can't say when it will be available, since OP adheres strictly to the "don't release it 'till it really works" rule, but I'd reckon it is safe to "hold your breath". The feature list covers everything you mentioned - including something very special on the gimbal control front. Also significant is that it will work equally well on fixed-wing and helis: meaning that, as an AP operator, you could cover all three platforms with one hw/sw base.

I'd be interested to see a similar thread on "what do you expect from your airframe/gimbal manufacturer?".

 

[Dewster]

Wow, nice list!.
* I think that the FAA may end up requiring a way to limit maximum altitude/range of a multicopter (for commercial use) if there is no telemetry installed. It would be nice if the manufactuers provide a way for users to define maximum range/altitude in their control settings.

* I'm a fan of "plug and fly" (provided that the plugs don't come loose and are secure). Ease of assembly/dissassembly is something that I like. It makes repairs simple (swap/exhange damaged component for one that works).

* And of course I fully agree with you on a multirotor being able to maintain flight on loss of a rotor/motor. These are things that would help the industry from becoming over regulated or viewed as being unsafe. People are wowed when I demonstrate some of the safety features of multirotor craft that I assembled; but they aren't enough...

The technology and know how is out there. I would like to thank the engineers for their hard work! Keep improving the technology!

 

[Bartman]

Good list!

I'm going to mention a name, but it should be obvious that I have no commercial motive - the OpenPilot Project has a similar (possibly longer) goal list with one or two very significant additions: like "lowest cost possible" and "open source". The (aptly named) OP Revolution was flying (in production hardware form) at the recent developers' meet in Portugal - think "CopterControl on steroids with full navigation and autonomy features". I can't say when it will be available, since OP adheres strictly to the "don't release it 'till it really works" rule, but I'd reckon it is safe to "hold your breath". The feature list covers everything you mentioned - including something very special on the gimbal control front. Also significant is that it will work equally well on fixed-wing and helis: meaning that, as an AP operator, you could cover all three platforms with one hw/sw base.

I'd be interested to see a similar thread on "what do you expect from your airframe/gimbal manufacturer?".

thanks for the additions. OP doesn't play nice in the sandbox though, i'm still a bit turned off by the program seeing how they swarmed this site and then bolted when people didn't praise their existence. the meltdown that followed with the one partner guy was also a turn-off. they may have a great product but if they want to have world acceptance they've got to thicken their skin a bit.

lowest cost possible is nice of course and "open source" is probably a very nice feature for someone with the mastery of software to take advantage of it. Manual PID tuning though, as I recall with the CC board, is a big barrier for adoption on a wide scale, IMHO.

 

[Bartman]

Flight control as we know it is all going to change. It will become highly configerable Data transfer. Like OSRC/ Ruling. Your receiver will do everything.

Isn't that what Mikrokopter is already doing with satellite/FC integration? Isn't the FC basically the receiver?

 

[Breezemont]

Isn't that what Mikrokopter is already doing with satellite/FC integration? Isn't the FC basically the receiver?

Flytron is too.

 

[jes1111]

thanks for the additions. OP doesn't play nice in the sandbox though, i'm still a bit turned off by the program seeing how they swarmed this site and then bolted when people didn't praise their existence. the meltdown that followed with the one partner guy was also a turn-off. they may have a great product but if they want to have world acceptance they've got to thicken their skin a bit.

lowest cost possible is nice of course and "open source" is probably a very nice feature for someone with the mastery of software to take advantage of it. Manual PID tuning though, as I recall with the CC board, is a big barrier for adoption on a wide scale, IMHO.

I think your view of OP is a little skewed . The core team has never participated "officially" in any forums other than their own. When the CopterControl was launched there was naturally an initial enthusiasm, with some participation here and on rcgroups.com - but the support demand escalated so rapidly that inevitably it wasn't long before they withdrew to their home turf. You're right that there were some "hurt feelings" because many people don't seem to "get" open source: at times the raw display of "entitlement" was galling in the extreme. OpenPilot is produced by a group of volunteers, working in their spare time. No pay, no profit. It's not a commercial venture that happens to publish its source code. So when some people bitched about availability or demanded extensive support or enhancements I think it's understandable that feathers would be ruffled. The "meltdown" you refer to was merely an example of the core project exercising robust defence of its own integrity - I'll say no more.

I think, too, you're missing one of the primary advantages of (successful) open source projects. If you, as a user (particularly a commercial one), invest in a particular platform then you run the risk of that supplier sinking without trace, dropping or replacing the product, or otherwise disappointing you. With open source, the rapid and complete world-wide dissipation of the intellectual property and expertise is often a far stronger guarantee of ongoing health and continuity of development than profit-motivated commercial ventures. In other words, even if you or I can't do much with the source code and circuit diagrams, there will always be plenty of others who can (and will), thus insuring our own position.

Manual PID tuning may well be more involved than simple "gains", but the advantages are manifold. "Gains" sacrifice flexibility and universal applicability for "one knob" ease of use. Look at DJI's embarrassing stumble: they optimised the gains for "normal" (i.e. small) craft only to find that they didn't work well on larger ones. In other words, gains are a gross simplification of the necessarily complex mathematics and, by definition, they mask/limit performance I take your point about widespread adoption, but OP doesn't have the profit motive that would justify such a compromise.

Returning to the question asked at the top of this thread - what none of us should want is the profit motive limiting, distorting and compromising the potential of this fascinating new technology. Examples of this are not hard to find, even this early in the game.

Just my 2 cents

 

[DennyR]

Isn't that what Mikrokopter is already doing with satellite/FC integration? Isn't the FC basically the receiver?

Bart
I was thinking more along the lines of No Tx. at all. You just have a laptop with a side stick for manual control. the stick controls the three main functions with one hand operation, plus a slider for the throttle. (Manual mode) the GPS data screen provides the real time waypoint info. which can be spilt to show the camera viewpoint. the camera movement can be done through toggle switches also on the stick. That is more or less where Dji are right now. You can have an endless array of auto modes.

 

[Breezemont]

It's not a commercial venture that happens to publish its source code. So when some people bitched about availability or demanded extensive support or enhancements I think it's understandable that feathers would be ruffled. The "meltdown" you refer to was merely an example of the core project exercising robust defence of its own integrity - I'll say no more.

This is where the hypocritical nature of the project rears it's ugly head. If they sell something, even at a loss, it is a commercial venture. Dankers can cry the blues all he wants about his out of pocket costs but as long as he is offering the product to the public for sale then it's not a hobby anymore - the loss of money at that point is purely the result of a dumb business model, not any altruism.
My experience with the OP team was that it's extremely cliquish and Dankers has very thin skin for a project leader. The CC zealots who stalked and climbed all over anyone who bought and resold a board ANYWHERE ON THE WEB made the whole project look like a total farce. You can't have your cake and eat it too, either profit doesn't matter or it does. I realize those people may or may not reflect the core group philosophy, but from the outside looking in it's impossible to tell who is the core group and who is not.

I think, too, you're missing one of the primary advantages of (successful) open source projects. If you, as a user (particularly a commercial one), invest in a particular platform then you run the risk of that supplier sinking without trace, dropping or replacing the product, or otherwise disappointing you.

You mean like OP dropping the CC board? It reached EOL faster than anything I've ever seen. Talk about a let down.
MultiWii, HVP, Quadrino, PiPo, Signguy, even Naze32 never invited me to leave their support threads because I didn't agree with the leader.
Just sayin'.

In other words, gains are a gross simplification of the necessarily complex mathematics and, by definition, they mask/limit performance

Yet Hoverfly seems to be doing quit well with their gross simplification, so is it a mistake to use a gain control versus PID? Or perhaps DJI just made a mistake in their implementation?

The overtone I get from OP defenders is almost always akin to the one I get from Prius drivers.
I "get" open source, I don't "get" Dankers' version of it.

 

[jes1111]

Who's "dropping" the CC board? It's being replaced by the CC3D simply because of component supply issues with the original. But it is plug-compatible with the original, runs the same firmware, to the same GCS, etc., i.e. it's a transparent change.

My experience with the OP team was...

Exactly

 

[DennyR]

This is where the hypocritical nature of the project rears it's ugly head. If they sell something, even at a loss, it is a commercial venture. Dankers can cry the blues all he wants about his out of pocket costs but as long as he is offering the product to the public for sale then it's not a hobby anymore - the loss of money at that point is purely the result of a dumb business model, not any altruism.
My experience with the OP team was that it's extremely cliquish and Dankers has very thin skin for a project leader. The CC zealots who stalked and climbed all over anyone who bought and resold a board ANYWHERE ON THE WEB made the whole project look like a total farce. You can't have your cake and eat it too, either profit doesn't matter or it does. I realize those people may or may not reflect the core group philosophy, but from the outside looking in it's impossible to tell who is the core group and who is not.

You mean like OP dropping the CC board? It reached EOL faster than anything I've ever seen. Talk about a let down.
MultiWii, HVP, Quadrino, PiPo, Signguy, even Naze32 never invited me to leave their support threads because I didn't agree with the leader.
Just sayin'.
Yet Hoverfly seems to be doing quit well with their gross simplification, so is it a mistake to use a gain control versus PID? Or perhaps DJI just made a mistake in their implementation?

The overtone I get from OP defenders is almost always akin to the one I get from Prius drivers.
I "get" open source, I don't "get" Dankers' version of it.

Breezemont You got right- it should have been spelled with a W.

I have had a lot of success with the CC board as a camera stabiliser but I refuse to add anything constructive at OP because of that person.

Here is a cantilever mount that works extremely well and saves a lot of weight. CC board was used direct to the camera. Zoomable long before Holger got there.
View attachment 4141View attachment 4142

 

[Breezemont]

Who's "dropping" the CC board? It's being replaced by the CC3D simply because of component supply issues with the original. But it is plug-compatible with the original, runs the same firmware, to the same GCS, etc., i.e. it's a transparent change.


Exactly

As I said.

 

[DKTek]

Wow, nice list!.
* I think that the FAA may end up requiring a way to limit maximum altitude/range of a multicopter (for commercial use) if there is no telemetry installed. It would be nice if the manufactuers provide a way for users to define maximum range/altitude in their control settings.

The technology and know how is out there. I would like to thank the engineers for their hard work! Keep improving the technology!

The MikroKopter controllers have done this and a lot more since day one of commercial release. http://www.mikrokopter.de/ucwiki/FrontPage .


For commercial use, the FAA is likely to require redundant systems and components. Something like this, http://www.micropilot.com/news-2010-nov-10.htm .
View attachment 4143

$$It's gonna cost a bit to play commercially$$....and be legal.

 

[Bartman]

I'm a bit turned off by MK at the moment.....I built an XY8 for a friend, it flew the best of everything I've built and it lasted all of about ten flight before blowing a BL. I'm very pissed off at the moment and very disappointed for his bad luck.

On the other hand, my XY8 with the HF board is continuing to come along just fine with very nice flying qualities and very competent camera control.

MK is nice when it works but there's a good chance that it will fail for no apparent reason. I don't think I'll offer MK build services like I had intended. Too much to worry about.

Bart