Help me build a Fixed Wing Mapping Drone

Maker

Member
Hello Everyone!

My goal is to build a Fixed Wing Drone for aerial mapping that would be capable to fly autonomous missions for at least 30 minutes at an average speed of 15 meters per second.

Such drone would probably consist of:

1) Pixhawk Autopilot (or any other autopilot that supports Mission Planner)
2) GoPro Hero 3+ Silver Camera (looking vertically down @ 90 deg.)
3) Emlid Reach RTK GPS
4) FPV system (preferred but optional)

Note, that since the plane will have a GoPro I think the power pack should be installed as a pusher.

Would you recommend FT VERSA WING SPEED BUILD KIT as a starter?

All the best!
 

Grakon

Member
Interesting!

Are you a Land Surveyor? I might be able to help you out. I along with my son are surveyors and we plan on building such a unit. I have not looked into the designs needed but I have several of of my own designs taken from original plans and converted to Flite Test style airplanes. We still may just build a quad copter it maybe the way to go.

Rich
 

makattack

Winter is coming
Moderator
Mentor
I have built such a beast, using the older APM hardware, as have others, but I might point out that for geo-surveying, having a camera that can encode GPS location as part of the EXIF metadata would be helpful. I'm not familiar with gopro cameras, but I don't think they do such a thing. If you're only interested in stitching photos together, you wouldn't necessarily need that, and can avoid using a "remote trigger" if you can just deal with putting the camera into an auto-capture mode with a set frequency. Might make the stitching/editing a little more painful, but it can be done.

I had built mine on a VersaWing with a blunt section, but rather than have the camera downward facing, it's just more fun for me to have it mounted for FPV use.
 

denis.nikolaev

Junior Member
Hi guys,

I'm working at Emlid, and I hope I can give you some useful tips.
Not sure, if I it will be ok here to just leave a link to our forum, so I am going to copy paste a useful info from our forum and then leave a reference link (I've checked the forum rules, it seems to be ok). Below you can find our CEO's clarification on using DGPS (e.g. Emlid Reach) for aerial mapping:
The main issue with precision georeferencing is that there is a random shutter delay and discretization of coordinates. Let's assume that we have perfect RTK solution in flight and are logging camera positions in Pixhawk as usual. After you have triggered the picture you do not know when the camera actually took it and you know a new coordinate at 5Hz rate. Assume that we are flying at 10m/s. 10/5 = 2 meters, between coordinate updates. Combined these two effect result in meters of errors even with RTK onboard. And if communication was lost in flight, that is another source of errors.

What Emlid proposes based on vast mapping experience?

There is only one way to know when shutter was actually opened, it is when camera attempts to fire the flash. This is why a hotshoe adapter (or internal modification for cameras with integrated flash) is required. You will need to connect Reach to the pin on the hotshoe and it will store timestamps of each picture. These timestamps are then stored in RINEX file along with raw GNSS observations. After the flight you will postprocess logs from base and rover in modified RTKLIB and it will give you a list of coordinates along with accuracy estimate for each of them. Postprocessing gives better precision and is more robust compared to RTK.

This method assumes:

  • No need for communication between rover and base, they log the data.
  • Not autopilot dependent, can work on Solo, DJI etc.
  • No autopilot software modification.
  • Requires flash sync cable to be connected to your camera of choice.

Some companies are sweet talking their customers into mapping with GoPro and referencing pictures with normal GPS. That simply shows lack of knowledge on their side. That might work for ok looking 3D models, but for precise ones that you can trust for measurements of volumes, areas and coordinates never. It surprises me how some are actually selling this as a business tool.

The quote above was taken from this post in our community forum. If you have any further questions, please drop me a line to denis.nikolaev@emlid.com

Best regards,
Denis
 

Maker

Member
Interesting!

Are you a Land Surveyor? I might be able to help you out. I along with my son are surveyors and we plan on building such a unit. I have not looked into the designs needed but I have several of of my own designs taken from original plans and converted to Flite Test style airplanes. We still may just build a quad copter it maybe the way to go.

Rich

Thank you for your post, Rich.

Yes, I am a Land Surveyor. It is obvious the way business is done in this industry is going to change very soon.

I am specifically targeted for a Fixed Wind drone and not a quad copter, since I already own two Iris+ and X8+.

All the best!
 

Maker

Member
I have built such a beast, using the older APM hardware, as have others, but I might point out that for geo-surveying, having a camera that can encode GPS location as part of the EXIF metadata would be helpful. I'm not familiar with gopro cameras, but I don't think they do such a thing. If you're only interested in stitching photos together, you wouldn't necessarily need that, and can avoid using a "remote trigger" if you can just deal with putting the camera into an auto-capture mode with a set frequency. Might make the stitching/editing a little more painful, but it can be done.

I had built mine on a VersaWing with a blunt section, but rather than have the camera downward facing, it's just more fun for me to have it mounted for FPV use.

I am so glad you posted in my thread! You are the one I need!

Let's forget about geo-surveying, GPS coordinates and EXIF metadata for a moment.

Let's concentrate on the craft, its mechanics and electronics to make it fly on auto pilot for at least 30 minutes at a decent speed (and I do understand that the speed very much depends on the weather).

So, let's make the list:

1) FT Versa Wing Speed Build Kit https://store.flitetest.com/ft-versa-wing-speed-build-kit/
2) Blunt nose Versa Conversion https://store.flitetest.com/blunt-nose-versa-conversion/
3) Power Pack?
4) Pushing Propeller?
5) Battery?
6) Transmitter GRAUPNER MZ-10 - 5 CHANNEL https://store.flitetest.com/graupner-mz-10-5-channel-transmitter/
7) Receiver?

Anything I am missing?

Please, advise.
 

earthsciteach

Moderator
Moderator
What flight duration are you looking for? (EDIT: Just saw 30 minutes) A Versa is fine for proof of concept, I suppose, but will be limited in flight time due to its size and will not be durable.
 

Maker

Member
Hi guys,

I'm working at Emlid, and I hope I can give you some useful tips.
...

Best regards,
Denis

It is very kind of you to post in here and render your support, Denis.
I'd prefer to discuss Reach RTK GPS related matters in Emlid forum (i'm registered there too), since RTK GPS might be of no interest to this community (though I might be wrong as well).

Besides, Emlid forum has more people specialized specifically in RTK GPS and can help find correct answers and solutions in a shorter time-frame.

I have noted your email and can also contact you directly, if required.

All the best!
 

epsilon

Active member
Yeah I'd use versa as poc but then maybe step it up to the kraken. I would think this will allow better flight in windy washer and obviously more payload capabilities.
 

Maker

Member
...A Versa is fine for proof of concept, I suppose, but will be limited in flight time due to its size and will not be durable.

Thanks for posting here!

Limited flight time due to the size? Less than 30 min? Should I go for a smaller craft or a bigger one?

Which craft would you then recommend to fly not less than 30 min on auto pilot with GoPro.

Regarding the "Durability", I believe you mean the foam frame. If so, then I am ready get 10 wings and switch to new one for every project :)

Please, share your opinion.

All the best!
 

makattack

Winter is coming
Moderator
Mentor
Here's the fellow who setup a similar POC based on the Versa wing, who inspired me to setup mine the same way:
http://forum.flitetest.com/showthre...rial-Photography-Platform&highlight=versa+apm

Mine is essentially the stock Versa Wing w/ Blunt nose and a motor/ESC similar to the "power pack C" from FliteTest:
https://store.flitetest.com/power-pack-c-fixed-wing-large/

I use a slightly different motor, the EMAX GT2210/11 Motor spinning a 9x4.7" prop setup as a pusher, 20A ESC, 4400mAh 3S 30C LiPo from ReadyMadeRC, FrSky X6R RX receiver (PWM setup), ImmersionRC 5.8GHz 600mW Video TX and RX (part of ReadyMadeRC 5.8GHz starter kit).

In this older video, you'll see the above setup using about 400mA over a ~5 minute flight. That includes some higher throttle runs up to about 75% -- so if you're easy on the throttle, I can see being able to run ~30 minutes with that pack. With how I was flying that day, 30 minutes would have used about 2400mA which is a little over half the pack, but I would never fly a pack down that low!
 

Maker

Member
Yeah I'd use versa as poc but then maybe step it up to the kraken. I would think this will allow better flight in windy washer and obviously more payload capabilities.

Hello! Thanks for posting!

Can the kraken be configured as a pusher?

All the best!
 

denis.nikolaev

Junior Member
It is very kind of you to post in here and render your support, Denis.
I'd prefer to discuss Reach RTK GPS related matters in Emlid forum (i'm registered there too), since RTK GPS might be of no interest to this community (though I might be wrong as well).

Besides, Emlid forum has more people specialized specifically in RTK GPS and can help find correct answers and solutions in a shorter time-frame.

I have noted your email and can also contact you directly, if required.

All the best!

Maybe I did not express myself clearly, sorry for that. My point was to put a little more attention on the camera choice, as not everyone knows, why it is crucial to choose a right camera, when it comes to aerial mapping (talking about sync output).

Thank you!
 

makattack

Winter is coming
Moderator
Mentor
The nice thing about the Kraken is that it's a twin motor tractor, so there's really no need for a pusher setup. The props can be made to be easily out of sight of any center mounted camera.

Tractors are more efficient as well, but the twin motor setup won't necessarily give long flight times. The other nice thing about the kraken is that it's designed for use with an undercarriage / landing gears, so that you can launch and takeoff using the autonomous launch/landing modes too, if you want to play with that.
 

mesolost

Junior Member
If you're doing the mapping camera straight down and you dont mind mounting the fpv camera on the wing since you know you'll be in mostly open area up high I'd suggest a storch, it can hold altitude with relatively little throttle, nice glide ratio, handles wind like a champ, can hold 2x2200 3 cell batteries and still has tremendous lift (inverted suffers with weight increase but mapping you wont go inverted) I think it could handle what you want. ^_^
 

Maker

Member
The nice thing about the Kraken is that it's a twin motor tractor, so there's really no need for a pusher setup. The props can be made to be easily out of sight of any center mounted camera.

Tractors are more efficient as well, but the twin motor setup won't necessarily give long flight times. The other nice thing about the kraken is that it's designed for use with an undercarriage / landing gears, so that you can launch and takeoff using the autonomous launch/landing modes too, if you want to play with that.

Thanks for posting.
I will consider stepping up to kraken in the future.

All the best!
 

Maker

Member
If you're doing the mapping camera straight down and you dont mind mounting the fpv camera on the wing since you know you'll be in mostly open area up high I'd suggest a storch, it can hold altitude with relatively little throttle, nice glide ratio, handles wind like a champ, can hold 2x2200 3 cell batteries and still has tremendous lift (inverted suffers with weight increase but mapping you wont go inverted) I think it could handle what you want. ^_^

Honestly I find the storch kinda fancy for a pure photogrammetry survey. Besides, it's transportation can be of another hassle.
Thanks for suggesting anyway. I wonder how many 2200 3 cell batteries can the Versa handle?

All the best!
 

makattack

Winter is coming
Moderator
Mentor
Considering I fly my versa with a 4400mAh 3S, I would say at least two given the right setup. I have also tried a larger, lower discharge rate battery to see if I could get a longer flight. Yes is the answer, but I didn't like the lack of punch. Barely could get it to loop, and it wasn't a pretty one.
 

Maker

Member
Maybe I did not express myself clearly, sorry for that. My point was to put a little more attention on the camera choice, as not everyone knows, why it is crucial to choose a right camera, when it comes to aerial mapping (talking about sync output).

Thank you!

I have used GoPro on Iris+ with regular GPS for aerial mapping? Of course I used Ground Control Points (GCPs).

I've read about the "Camera timestamp integration" following your link to Emlid forum and have found the information very useful. However, I am a bit confused on the topic, could be due to lack of my knowledge.

Let's discuss the figure below:

GoPro.jpg

Let's assume i have an auto pilot mission with 4 Waypoints, and these Waypoints are at equal distance from each other. Having the waypoints at the same distance (grid type) does not guarantee that it will take UAV same time interval to reach each of them. It means that UAV may reach W1 in 3 seconds and W2 in 3.5 or 4 seconds, etc. Photos taken from the Waypoints must indeed be flash-trigger to sync with centimeter accuracy. It is great to see that Emlid team has come with this idea and implemented it.

GoPro time-lapsed based photos are taken at exactly same time interval (every 0.5, 1, 3, 5, 10 seconds) REGARDLESS whether or not UAV has reached the Waypoint. So if Reach Rover logs its GPS locations throughout the flight in a suitable time interval then it is only a matter of syncing the first photo, the rest will fit as well.

GPS time is very accurate and my research over internet proved GoPro's time is also accurate enough for syncing (http://lumberjacksystem.com/get-accurate-time.html).

"Any GoPro with WiFi can be used as a very accurate time source."

Now, I think, if Emlid team can implement a function inside Reach to adjust and sync time on GoPro Camera over WiFi before the flight, then we would end up with a perfect match between Reach GPS Logs and GoPro Photos.

Please, share your opinion.

All the best!
 

denis.nikolaev

Junior Member
GPS time is very accurate and my research over internet proved GoPro's time is also accurate enough for syncing (http://lumberjacksystem.com/get-accurate-time.html).

"Any GoPro with WiFi can be used as a very accurate time source."

Now, I think, if Emlid team can implement a function inside Reach to adjust and sync time on GoPro Camera over WiFi before the flight, then we would end up with a perfect match between Reach GPS Logs and GoPro Photos.

Please, share your opinion.

All the best!

All cameras have their inner delay, different for every picture, this is why an external trigger sync is not reliable. I think this is the same case with GoPro, its inner clock can be great, but we can't determine, how much it takes for the camera to take a picture. 1 ms delay can lead to offsets of cm's. On the other hand, it may be ok, depending on the accuracy you are expecting to receive.

Regards