Why can't we have the best of both worlds in FC's?

[DKTek]

Ok, so I've been an MK guy for a long time. I absolutely love the constant updates and cool features set. The telemetry is awesome and the Navi log files too. But, the system as a whole is sloppy at holding position and altitude. It's greatly affected by the elements, making for very inconsistent flight performance. This frustration led my partner and I to other FC's.

Our friends and competitors use DJI so we built a new Okto with the WooKong-M. It's ability to hold position and altitude seem second to none. Aside from the occasional freak out with the GPS mode on, it's pretty much rock solid. The kind of performance expected for carrying cameras and sensors. The elements seem to not bother this system either. Only real issue is the lack of useful telemetry features.

So, why can't we get a FC with the best of these two FC's? I want all of the features of the MK and the stability of the DJI. Is that really too much to ask for? Also, would it really cost that much more for the boards to be populated with certified components and parts? Something that may some day get FAA certification for commercial use. The way I see it, we're headed that way anyway so why wait?

 

[Bartman]

The more things change the more they stay the same! By your member date you must know it's been like this from the beginning! We tried to establish a flight controller assessment that organized features but it was depressing to look at as it just accentuated how bad the problem is!

I mentioned in another thread, have you ever tried putting the GPS Gain on a slider on you radio so you could adjust it? What works for this setting on a calm day isn't good for a windy day and vice versa. I can't imagine trying to fly Mikrokopter without having this adjustment available from the radio.

 

[DKTek]

Hey Bart, I do in fact use a slider as well. You give good advice here!

My member date is exactly why I ask this question. Hasn't it been long enough? People like Holger and DJI have the combined skill set to bring the hobby grade FC's to a much higher level. They know how to make the equipment, just use better parts. Something a little more robust.

On another note, the current lineup of hobby grade equipment needs to follow what's going on here. There's a commercial explosion about to happen and they aren't ready. Those operating under a 333, IMO, are doing so only to temporarily get Congress off the FAA's back. Look at the limitations imposed...closed sets, middle of nowhere, Etc. Pretty safe huh? The closest FC to any certification is the MicroPilot and even they do not use FAA approved parts. Sure, they are MIL spec'd for Government/Military use but that is not the same as the FAA's standards for full scale aircraft electronics. The MicroPilot has triple redundant FC's but they are all mounted on a common board so there really isn't any redundancy in their high end product.

 

[Bartman]

I agree that hardware should have to pass some well defined set of standards but if I had to choose at the moment I'd subject the firmware to testing first! How much money have we lost as a community of volunteer and disregarded beta testers? Firmware testing would go a long way to verify failsafe algorithms, GPS signal handling, and the limits of what the FC's can do with extreme/complex flight demands.

we've come so far and have such a long way to go!

 

[Old Man]

The closest FC to any certification is the MicroPilot and even they do not use FAA approved parts. Sure, they are MIL spec'd for Government/Military use but that is not the same as the FAA's standards for full scale aircraft electronics. The MicroPilot has triple redundant FC's but they are all mounted on a common board so there really isn't any redundancy in their high end product.

Actually, for the most part MilSpec'd stuff that's in use in military aviation applications is viewed the same as "FAA certified" equipment. Both are listed as approved under the FAR's, which is how corporate aerospace gets to fly a lot of new technology without having to first go through the FAA-PMA certification process. They just install it in a military aircraft and it's instantly covered. Of course it's pretty well tested out first. Taking that a step further, the FAA doesn't really certify equipment. They leave the certification process to private industry and engineering and accept their results as the foundation of the part certification process. The FAA reviews and approves, contraction to outside agencies like ASME to vet the presented data.

 

[Old Man]

I agree that hardware should have to pass some well defined set of standards but if I had to choose at the moment I'd subject the firmware to testing first! How much money have we lost as a community of volunteer and disregarded beta testers? Firmware testing would go a long way to verify failsafe algorithms, GPS signal handling, and the limits of what the FC's can do with extreme/complex flight demands.

we've come so far and have such a long way to go!

Bart,

Is this not exactly what APM/Pixhawk has been doing with the open source software? Much of what DRTek is wishing for is largely already available, just not as easy to use as it could be.

 

[Bartman]

easy is always the last feature to be added!

i'm ready to try a Pixhawk, I've got everything else ready to go, need a heavy lift backup for my MK/Movi rig

 

[DKTek]

Hello Old Man. I get your point but try rigging out your private aircraft with a full complement of military gear and see if it'll pass an FAA inspection. They are the determining factor here.

We do have two pixhawks. Open up the case and you'll still only find average components. The software changes at will but no matter how good it may seem, it still relies on the hardware. Software can sometimes work around poor hardware but why should it. Give the ever changing software a better foundation to begin with. If it's for commercial operations then heck yeah it's going to cost a lot more.

 

[Old Man]

I'll go with you on that but a lot of the UAV stuff being sold to the military is still filled with common components. Heck, I've seen the made in China stamp on quite a few of those electrical components. I will agree that MilSpec hardware is significantly more robust, reliable, and accurate than common items but the cost goes up with the spec. I don't know how we'll get "affordable" and "certified" to associate in the same sentence. The "certified" part typically eliminates the "affordable" side by the additional costs levied through the certification process and the liability that has to be borne by the certification agency.

 

[SleepyC]

Old Man... you DA MAN.. your knowledge of all of this is inspiring to say the least!

 

[DKTek]

Old Man, you are exactly correct! But we are not the military or even a public service agency. They are not required to adhere to FAA standards. That's right people, your local police helicopter is not required to adhere to the same regs that we are. Don't get me wrong here as they do generally follow FAA standards for commonality and safety, but they are not required to.

Old Mans statement on the certification process is dead on. So, what are we looking at here? Maybe a different catagory of cert process? Probably! What do we have to show from a hobbiest product side? A long list of failures...fly aways, not so good QC of parts shipped, mostly Frankensteined drones(although DJI is changing that), in flight failure of motors, props, etc....hopefully you get the idea here, but if not... Look at it from the FAA's view, our FC's and associated gear kinda sux.

This is why I say the hobby industry is not ready for commercial use and has done little to prepare for the obvious. It has been obvious for long enough of time that they should have by now had a robust and reliable system available for the "PRO" , even at an expected cost increase. At least show the FAA some kind of real progress by stepping up the manufactures game.

So again I ask, why not have the best of both worlds? It's not that the manufactures can't give us that! Less time on the Bling, more time suring up what you already have. Do us all a favor and consider this...

 

[Old Man]

Thanks, but I'm just like anyone else, just been doing different things. What I really like about this forum is we all learn from the knowledge and experience of the others. All you guys have experience up in your heads I don't have any inkling of yet. Still a lot to learn for me too, and I get to do that in the company of adults. How great is that?

 

[dazzab]

The developers of Arducopter/Arduplane are a very impressive group of people that are highly qualified. I think they would be having very similar discussions about components. 3DR appears to be very progressive and I would be willing to put money on them improving the hardware of the Pixhawk in the not too distant future.

 

[Old Man]

I think Dazzab is correct in where our "certified" flight systems will be coming from. Eagle Tree will be another that ends up doing similar, especially since they already provide a considerable line of sensory equipment that's the foundation of what is currently used in several sUAS systems. That 3dR is already listing their products with an ITAR identification number suggests preparation for deeper technical involvement. I'm surprised companies like Cloud Cap haven't jumped all over this but I know Lockheed and a Swedish outfit have already developed a couple of FC's that will meet avionics certification standards. The price point jumps up by a factor of 20 or 30 though. The number of common RC modeling manufacturers that currently provide parts for high end military systems is simply amazing. Futaba, JR, 3w, DA, APC, Mejzlik, and Sullivan instantly come to mind. It's how they are intermixed that makes things work really well.

 

[DKTek]

Case in point here...

http://www.washingtonpost.com/world...d8a07a-86c2-11e4-a702-fa31ff4ae98e_story.html

 

[Old Man]

I read that, and once again it demonstrates the lack of knowledge some of the safety inspectors have. They have just about completely failed to consider maximum range, flight duration, or consider the remote area/confined area requirements of the temporary COA's. So their thought processes are severely constrained by what they knew of full scale operations, not low level, limited range stuff like we fly. What's the average flight time of a heavy camera cinema rig? 7, 8, 10 minutes? How much of that time is spent in pre and post take off checks?

I agree we have problems with some of our flight systems. One maker seems to be considerably more of a problem than others, yet that's the same company that's incessantly marketing more and more sophisticated systems at higher and higher pricing. The speed at which their new products are reaching the market implies that less than complete systems testing is being performed before being released to the public for unrestricted use. Yet people keep buying them because they are easy. The stuff that takes a little time and work to learn appears to be far more reliable and have more user set failsafes than the buy and fly stuff. I suppose that takes me back to some regulatory suggestions that irritated a lot of people, requiring that people become knowledgeable of their systems and how they function before entering into commercial flight operations.

I think we can indeed have flight control systems as good as what the military uses for Tier I and II sUAS systems. Even some Tier III. In fact, we already do, and I'm sorry to say that DJI systems also fall into that category. We have much better, and for less cost even, but the whining starts when someone says one has to have a level of personal involvement for them to function as well as they could. What you don't know is the military Tier I, II, and III systems experience all kinds of flight control failures in one form or another...but those crashes or fly aways (it does happen) never make the news headlines. That information is withheld from the media and public because most all of it occurs in war time environments in distant lands. Loss rates for some of them have neared 80%-85% in the recent past. They're getting better but not enough to say they are more reliable than what we have. Pretty scary when you think that's the stuff that has lobbied hard for opening our airspace. Things like Predator and Global Hawk are a lot more reliable and robust but for many millions of $$ per copy they damn well better be.

At the moment we are at the mercy of a well coordinated media ploy intended to make the people of the U.S. DEMAND our stuff be heavily regulated, or at best not protest when heavy handed regulations are imposed. Without cause they cannot hope for effect, and they are getting maximum mileage from any and every minor incident involving a citizen operated "drone". Wish they did that with situations that actually kill people, like drunk drivers... Remember the word "drone" is associated with the killer attack drones like the predator and reaper where the public is concerned. A Phantom is not what pops into their mind when they see or hear "drone", and a cinema rig is something that is far beyond their wildest imagination.

 

[DKTek]

Looks like it was an S1000. We've had a few uncommanded fly offs with DJI in GPS mode but they didn't get far because we simply clicked over to manual control. It's a simple procedural item that should always be part of the pretakeoff checklist. It's a responsible mental thing and as pilots of full scale aircraft, emergency procedures should always be a consideration during every phase of operation.

Now, I do not claim to know what happened on that flight but I do know that in most cases when a DJI GPS failure happens that the craft will continue to fly off if allowed to remain in GPS mode. IMO, that is why so many fly-offs end badly and DJI gets a bad reputation. If the pilot has peice of mind to click back to manual control immediately then the flight is not lost.

Problem is that these systems are too easy for anyone to fly in GPS attitude mode. Seems that far too many operators can't fly an RC helicopter to begin with and would crash anyway if in manual mode. Maybe part of the license test should be to demonstrate full manual mode control both flying and hovering. For commercial use nose in is a must along with unusual attitude recovery. Just demonstrate a competent set of flying skills similar to advanced heli competition.

Seriously, how many multi rotor owners flew helis prior to getting into MR's? How many actually know their machines and all of its functions and capabilities? I'll bet the ratios are far worse for Phantom owners...and any other turn key MR's. Sorry DJI, anyone can fly your MR's but it takes heli skills and competence to OPERATE one!

 

[Old Man]

I'll have to admit that my ability to fly a regular helicopter usually ends up a disaster, but I can fly pretty much any size RC plane pretty darn well. I've tried quite a bit over the years to get into regular helis but just didn't spend as much time with them as I could/should have to get good with them. My focus was more towards giant scale fixed wing and the division of attention didn't fit what I wanted to do.

Your comments take me back to some other regulatory requirements posed for those that have a desire to fly professionally, having to demonstrate knowledge of the aircraft systems, flight controls, and emergency procedures. I'm certain that some became uncomfortable with that but if someone wants to be a pro they should be held to a higher standard. I don't know what to do about the hobby level people, and tomorrow will likely make many of us wish we did. The thought of so many new users opening up what will be Pandora's box to get at their new toys is a little disturbing. Perhaps eliminate GPS slaved flight modes from hobby level aircraft but that will gut sales of multirotors at the hobby level.

I don't want to get into a DJI FC debate because I've already been there and won't ever use one again myself. There's definitely something that needs fixin' before they move any more products to market but that hasn't happened all year and doesn't look to be getting handled any time soon, making a hard case for those using the stuff to be fully cognizant of how they function, what they are prone to do, and how to deal with a situation instantly if one becomes evident.

 

[Old Man]

Here's something to think about as things move forward, we're all going to be facing what's known as "Product Liability". The Chinese makers don't have any skin in that game because they know courts can't touch them. They hold too much of the world's debt for any government other than their own to cause them any grief. But each dealer and user of a product is going to be held liable for damages caused by aircraft using components with known or perceived defects. In the U.S. there's a couple of makers of flight controllers that have been going the extra mile to assure faithful reliability, likely for reasons of minimizing product liability and of course to capture a larger market share when the regulatory hammer comes down.

As a commercial operator the lawyer for the injured or deceased isn't going to accept an explanation that an aircraft flew away and caused harm for an unknown reason. "Chit happens" just isn't going to cut it. If there was previous history of that product doing something irregular or uncommanded the commercial user will be charged with criminal negligence. As we start insuring our aircraft for liability and hull coverage we'll find the cost of coverage will escalate with the type of product being used. Enough loss generating incidents and coverage will become unobtainable, putting the user out of business. If enough losses occur denial of coverage could spread across all brands of products in use for that purpose, shutting down an entire market. This is especially true if regs are passed that require liability coverage for commercial ops.

So, do we put the onus of product functionality and reliability 100% on the manufacturers? Somehow I don't think that will work. I believe many of us think a lot of issues have been caused by operator ignorance, which the maker can't be responsible for unless they failed to provide user manuals. Just like anything else that's used to make money users will be expected to understand and know how what they use works. You can't drive a car or truck without understanding the basic equipment so why should that be different with our aircraft? It comes down to responsibility. Some embrace it while others spend a life time avoiding it. Those that believe they can run to a hobby shop and buy a camera fitted RTF by anyone copter and instantly start an aerial imagery business will continue to be a problem as some of them are now. Some will crash their new toy and lose interest, while a few will decide to invest the time and effort to truly understand. It's people like those in this forum that have an opportunity to protect their future by reaching out to those new people they see flying irresponsibly and assisting with their education. The pros can make contact with manufacturers with suggestions of how products can be improved to increase safety and reliability. With a little work those same people could put together a code of standards and ethics that could become the foundation of trade craft.

 

[dazzab]

If there was previous history of that product doing something irregular or uncommanded the commercial user will be charged with criminal negligence. As we start insuring our aircraft for liability and hull coverage we'll find the cost of coverage will escalate with the type of product being used. Enough loss generating incidents and coverage will become unobtainable, putting the user out of business. If enough losses occur denial of coverage could spread across all brands of products in use for that purpose, shutting down an entire market. This is especially true if regs are passed that require liability coverage for commercial ops.

This is exactly why I can't believe people don't see the very serious ramifications of a case here in Australia where it was proved in a lab that DJI flight controllers are susceptible to radio interference and can lock out the operator for up to six seconds to prevent false failsafes. Why in the world would such a dangerous product be allowed to be sold and used around the public? It's only going to take one legal case such as you describe to wipe out DJIs market so you would think that they would address these issues and start listening to their customers. DJI needs to have it's code independently reviewed just as every flight controller should.