Hi,
Open source autopilots are widely used in the RC toys. I tried 3DR's pixhawk. It seems to be quite okay. There are also other open source autopilots.
The commercial autopilots are always used in commercial unmanned aircraft. They are somehow much more expensive than open source. I’m wondering what are commercial autopilots’ typical advantages compared with the best open source autopilots for conventional fixed wing usage. Could anyone give any hint on that?
Best,
Anna
Replies
After programming and flying over 500 autopilot missions with an FX-61 APM2.5 set-up I strongly disagree with John. During these 500 missions there was only one mechanical failure (elevon servo popped out of its housing) , ESC failure once and one failed mission due to solar flare activity. ALso had a folding prop fail to engage after gliding in a crosswind twice. Then had a few failsafe events not properly set that caused problems.
I also use APM2.5s for other projects were I have them turned on for weeks at a time. They are robust if properly cared for and kept dry. If given too high a voltage or dirty power supply or you use ESC for power supply you are asking for trouble.
The APM is fragile, but also robust if given a clean power supply and protected. and there are clearly varying degrees of quality assurance per manufacturer.
Here are the main failure modes I've seen from 500+ flights.
1) programming error. Obstruction between waypoints not accounted for. Don't RTL if UAV is behind a tree.
2) boot-up GPS altitude wrong and if using Abs-Alt mode for mission you can have severe altitude errors with auto-launch. Reboot APM again until GPS altitude is in line with mission planner Alt.
3) launch waypoint too close to launch point...UAV does a 180 and comes straight back. If you are auto-launching in a tree line corridoor you will have a collision.
4) Flap/horn/servo Detachment. Make sure these parts are of the highest quality
5) Esc failure...use backup ESC and highest quality
6)If you use a folding prop, don't ever glide. It will fold up in a cross-wind and get bound when the UAV throttles back up
7) high winds will flip the UAV. If APM is not correctly programmed it will not re-invert UAV. It will crash.
8) Fail-safe issue causing UAV to RTL at a place were there was obstruction between RTL command and launch point.
9) Don't use GPS for altitude, especially if auto-launching. Your plane will fly erratically as the GPS alt algorithm struggles to calculate the correct height.
10) put packing tape or something to reinforce flap attachment to wing
11) When plugging in micro USB, be gentle. Those are micro soldering joints so if you wiggle it the APM will reset etc
12) if the UAV wags its wings or porpoises then tune it properly otherwise this will stress the airframe and cause eventual failure.
13) don't fly on auto-pilot mode if there is high solar activity. This will bring your UAV down
14) Don't launch in tail wind unless your UAV is highly powered.
I always use auto-launch and auto-land This helps minimize failures.
So my long winded point is....the APM itself is robust...if you power it up it will work for weeks on end. The failure points are generally elsewhere and they can be minimized with high quality replacement parts and diligent Mission Planning.
John, do you work in the military or use military grade UAVs ?.
When flying over populated areas it would be wise to use a military grade UAV though. They still crash though just like the hobby grade ones.
Wow FM,
That WAS long winded :-)
But I agree,mostly........
Other than "Auto Launch/Land", if you can't fly manually, you have no place in the "small" UAV market.
It will ALWAYS take a certain amount of skill, and understanding.
As the saying goes, it is impossible to make something foolproof, because sooner or later, you WILL run into a sufficiently talented fool :-)
Pixhawk is truely the next gen autopilot with hundreds of hours of test flights and a community with 62,000+ members.
Consider it with this simple example.
Easton La Chappalle , a 17 year old boy has made a prosthetic arm under $500 that costs $80,000.
Anna, If you want stadistics, aprox 10 APM at my club, 3 years flying, anyone fails the hardware in tree years, 3DR and clones, I had two crashes at the begining, for bad configuration and no reaction in time (little experience); any drone lost with APM yet at the club, one expensive drone with Naza flies to the sea, didn't respond and lost with an expensive camera, four Phantom's reported lost the last six months in our local forum with Naza, a very expensive airplane with a 20.000uss tetracam and mikropilot touch the ground at 250kms/h, I saw the photos, the bigger part had 10cms, I know another big plane with mikropilot but I never saw flying, in the ground no problem, I was reported that two expensive sensefly crashes in the north of my country and very few flying here, my conclusion, it's no sense to pay expensive equipment, they crash the same than a cheaper one, if you only likes to fly choose a Naza, isn't so expensive and it's easy to fly; if you want more, know how your autopilot works, continuously new capabilities to test and the possibility to discuss with developers and other users, recive experts help to build your own Drone, and make a lot of friends, APM, Pikhawk is the choice.
Sensefly is a very overpriced, gimmick. They use hobby grade material. Their customized open source Mission Planner locks out the cool features like "Fly Here" , Follow Me, One way flights. Their failsafe settings are locked out so when the SenseFly goes outside telemetry range it comes back and you can't ever change this. The airframe is small and has limited range and payload capability. The poly-carbon frame looks nice but explodes in crashes. The point cloud features they charge an arm and a leg for are easily made with free software like Photosynth etc. If you buy this product (or Ebee) you will have regrets. Their sales pitch is clever indeed but what they rely on most to sell their product is gullibility of the consumer.
Greetings,
Thank you for your observations on the Sensefly. I was very frustrated by their $25,000 price tag and forced inclusion of Pix4D software. For that much money, you can buy two airframe/sensor setups and processing software and still have some money left over.
Any suggestions for other commercial solutions that are using better components and that do not bundle Pix4D?
-David
Agreed, except that we managed to get the same capability for 1/10 of that price tag... (OK, we are fine with an EDU licence of SFM software as an university using our UAVs for research ... but still).
On the other hand, within the professional surveying/remote sensing community the eBee is considered to be the cheaper/cheapest option... as far as I was able to perceive from attending a few of their conferences... And once you start comparing it to GateWing/Trimble and alike, you actually might start to understand their point...
So to answer your question, if you want an out of the box commercial/"professional" solution for surveying, I believe you will actually have hard time to find anything cheaper.
Yes, as expensive as the eBee is, many companies understand that time is money. If it is the cheapest solution to get an RTF UAV system, then it's worth it.
I am surprised that nobody has stepped up to provide a competitive system, say a Pixhawk on an FX-61, for $5-10k. 3DR?
Greetings,
Part of the expense in the eBee (and the 3DR) system is the mandatory inclusion of Pix4D. You need to buy this even if you have another solution, and even if you already own Pix4D. Sensefly and 3DR could both be more competitive if they unbundled the software.
Take a look at Aeromao for a less expensive option.
http://www.aeromao.com/
-David
Hi Anna,
I'm also rather curious about your intentions for the autopilot.
If you were to look at the last "Outback Challenge", it would be clear that PixHawk would be the obvious choice, even if you lose 25% of your planes, compared to commercial systems.
And anyway, the price of a commercial system normally, or should, include operator training.
I fly APM 2.5/6, in both Quad Copters,. and Fixed wing, and the only three crashes I ever had, was caused by pilot error.
Then, I fly non commercial, hobby missions. but so much for product reliability........
There is NO system, open source or commercial, that you can buy, plonk into an unknown airframe, and get it to fly perfectly the first flight.
Just my 2cents......
-
6
-
7
-
8
-
9
-
10
of 15 Next