First of all some background...
Having spent a good 2 years playing about with an Skywalker, Ursus and a Canon camera running CHDK, and designing my own uav, which actually flew recently ,I have come to the conclusion that flying wings may just be the best type of platform for short range aerial mapping tasks to create DEM, DSM models for GIS purposes.
The Skywalker was okay (but quite fragile, and now in pieces!), but I think it is too small to fit a roll gimbal on it. And I've found that if pictures are taken on a slight angle, it is very difficult to use software to tile them up nicely. this is my best effort, but far from perfect. The roads on the right do not line up as the camera must have been on a 45 degree angle when the picture was taken. (click for larger resolution) - It was taken using an ixus80 attached to an URSUS Airframe
Realizing that I need to use an airframe that can accommodate a stabilized roll gimbal, i started to design and build my own twin boom design. It took 6 months from initial design, to first test flight. It has a wingspan of 2.4m - AUW of 5kg, and a flight duration of around 30 minutes with 2 x 5000mah packs. However it needs a heck of a lot of room to take off and land, something i don't have in the areas i need to take photos.
Picture of twin boom plane...
What are peoples experiences with using flying wings for this reason? I have the following types of flying wings, and wondering which one is the best to begin with?
I have been using the 60" wing to test the APM2 and 3dr radios. And was using the URSUS (now in pieces after a nasty crash!) as a test platform to practice with the Canon CHDK Kit.
Here's a few pictures of the wing and camera....(click images for larger version.)
My question is how straight forward do you think it is to fit a roll stabilized gimbal to any of the above airframes?
Hi Coby, I've just picked up your posts, and will be doing another few flights tomorrow morning, I'm going to use Telemetry to test the performance. I am looking at a better off the shelf wing design, as my scaled up Zagis don't handle wind that great.
I looked at adding a pod under the wing early on in this thread - I also posted some drawings, If i need a gimbal i will look at doing a pod of some sort,
I am going to be building a Catapult launcher this weekend, as I am finding it difficult to launch the wing, and the 300W 10 x 7 prop is ever so close to my fingers...!
I did not see an answer to your earlier question about CG being measured from LE or TE - It's measure from LE @ center line (nose). That calculator is a handy piece of gear.
You could be seeing more issues in winds with your scaled up Zagi's since you probably increased aspect ratio and reduced wing loading. Good for endurance but perhaps not so much for gust sensitivity. I am guessing that is why you see so many small commercial wings around 1m span (besides affordability and portability). Load up the wing and fly a bit faster and it will penetrate like a lawn dart. Not sure what kind of speed they average. We have had the Nurf out in 25mph winds and it did OK, but could not tell you how it would do with a still camera. And we fly all R/C so apples/oranges right now.
Your camera FPS, FOV, and AGL altitude of the plane would probably bound your speed range for best image capture with proper overlap but that is not my area. Assuming that is true then you need to find a plane that has a max range in the same airspeed range that your camera can capture good data. Wish I had some concrete numbers but that's all on my list to do. I am sure the photogrammetry pros on this thread can comment better than I.
I think you could keep it simple with the launch and use a bungee. Just need a small hook on the bottom of your wing a couple inches forward of CG (we use hooks from the hardware store around here). Be nice to have the hook mounted in ply reinforcement and well glued/taped in. The other benefit of the pod is it could give you something to hold onto for hand launch. I understand about having that prop buzz by your fingers. Better if you can launch and then throttle up when able.
Hi Coby, Cheers for the info RE: CG. I found it some where whilst trawling the internet, I thought that it would be easy to find that kind of info!
I am seeing issues with stability with the scaled up Zagi, I think the main problem is that i have not scaled up the wing tip vertical stabilisers, and the plane is yawing quite a lot. I am going to double the size of them for tomorrows flight. I do have a regular sized Zagi that i may use if i can't get the large one to fly properly.
There is an excellent guide on the internet written by the chap from pteryx uav. it's called 'the secrets of aerial mapping' or something like that. It has lots of great info on there regarding aerial mapping, graphs etc.
I already ordered all the parts necessary to make this launcher catapult. I had 4 crashes on launch today as there was no wind and hand launching was difficult. I am launching on Stabilise, as i don't trust the APM take off feature to start the engine after the plane has left my hand! I will be using Auto take off with the catapult though.
I made some good progress today and had great weather and the opportunity to drain 4 x 2200mah batts. Did about 6 flights by my local field, the pictures are great, sharp, good colour, however most of them are taken at an oblique angle, and Microsoft ICE and Autopano Giga are unable to do anything with them.
Therefore I am going to be building the Roll gimbal this week. That together with the catapult should be the last few pieces of the puzzle for me to start mapping....4
At least it's summer and my local field isn't waterlogged anymore! so plenty of flights will be happening!
I did 4 flights yesterday, The first flights 1 & 2 were just testing and tweaking camera settings, nothing to report there.
The Second 2 flights are shown below, wind was 9mph. There was sunshine and a bit of cloud cover at times
The 3rd flight was a cross wind flight to see how well the wing would handle in a light crosswind, the answer is not very well! Crosswind flights are not good!
The 4th flight went much better, the results are shown below,
Way-point 1 is in blue as i had to switch to stabilize quickly, hence the difference in colour. this flight was much better and i am happy with the wings performance, It was using 65% throttle and maintained a constant 100m altitude for the whole flight.
All of the photos came out very crisp indeed, they are a bit light though, I used 1/1500 shutter speed on ISO 200. Probably should have dropped ISO down to 200 maybe. Resolution: 3264px x 2448px.
Here is a sample photo from the flight..
ICE and Photosynth really struggled with these images, and tried to stitch them as best they could, but the results are awful!
I did manually remove all the oblique shots during the turns, but there are still a few semi-oblique shots in there. Also, being a wing and having no rudder to keep it on course means it needs to bank to keep it on the flight-path. Therefore I think the
Next steps are to add a simple roll gimbal to ensure vertical shots are achieved.
onwards and upwards!
It looks like you tuned the APM nicely by the 4th flight. I'm not sure why ICE is giving you such a hard time, other than the reason you mentioned about trouble matching. Can you boost contrast a bit to try to accentuate some of the patterns in the imagery? I am investigating opensource photogrammetry software and will keep you in the loop on what I find.
I'm sorry about the bad link last time I posted here. I meant to provide a link to my SW UAV CHDK camera setup here: http://www.youtube.com/watch?v=01iAJkiFAX0&list=UUNVz08pPZ2XTbA...
And since then I was able to get some help in post processing some pictures from the flight I was referring to and I am very excited by the results. The gentleman who did the post processing for me, Mark Willis, has access to high end photogrammetry software, leica suite. My images look comparable to the sample one you posted above in many cases in terms of angle though the sw tends to stay level and fly sideways in wind rather than rolling which gives an advantage. I gave him a small subset and I'm very happy with the results. Here is a quick video of a resulting orthophoto and DTM he put together for me to ensure me that the images I am capturing are good enough for post processing.This was in 12-16MPH winds on my skywalker shown in the video above. No gimbal used.
I am determined to figure out a low/no cost solution for post processing. I am just starting that exploration process and will post results as I make progress.
Great job on your flights!
HI Brendon, (spent a while typing a detailed response but lost it!!!! Grr.) first of, great job on the videos, both the videos are great, I like the Canon 260, but i think i'm going to experiment with a 230 as i am still not that confident in the APM2 yet. The DEM you created is very well done, I understand that Leica is the industry standard and very costly.
I have been playing around with the software Microsoft ICE just to try stitching the image set i took yesterday, here are my slightly better attempts, still a bit dodgy, but getting better...The main problem is that the photos are duplicated, i need to ensure i take a single flight and run it through the software....it doesn't help that there is no detail in the fields!
This week i'll be spending most of my time building a catapult and a roll gimbal...i'll probably end up with another Skywalker! - but i do like the durability of the wings.. Next weekend i'm hoping to do some flights at some sand dunes.. The software should cope better, and produce nice DEM models as well.
Try Photosynth and meshlab for free photogrammetry software..
Hello, Richard! Could you send me an archive of all pictures from one of your flights?
I've obtained a trial version of a very advanced photogrammetry software and have access to a bank of powerful computers. I can make several orthophotos and DRMs for you! Both my planes are broken and the trial will run out before I get them back together.
Hi Kirill, that sounds great! it would be interesting to see your results...sent PM.
After removing more photos from the flight - I have created a slightly better mosiac - not perfect but good enough considering the subject matter. I have definitely learnt to only fly over an area once! multiple images of the same subject matter may be great for a point cloud, but no good at all for a mosaic, unless they are geotagged of course.
Time to try and tag the pics with the tlog.
Lets see if i can improve the below image some more.
Longer runs less turning take the images from one direction only for stitching (normally the into wind ones) You have some great images there. Chuck them into photosynth and see what happens.
I will do some longer runs next time. I did try and plan the flight path so that all the turns would take place outside of the large square field in the middle of the image. There was a road in the bottom of the image which i did not fly over. Hence the oblique trees.
Is there any reason why it's best to use images from a single direction?
I did upload all 700 of the images into photosynth from all 6 flights, with multiple images of the same areas. Here are the results, I also downloaded the pointcloud and imported it into meshlab. Unfortunately, the results are not terribly interesting due to the flat nature of the ground. There is however an interesting spot of relief in the light green patch in the upper right had area of the image.
I am in the process of trying to GeoTag the image using the APM planner tool. The .tlog is 10mb and includes all of the flights. Totalling about 1 hour. I am only interested in Geotagging flight 6. about a 10 minute flight at the end of the .tlog. If i am interested in the data from 50 minutes onwards, do i put 50 minutes in the seconds offset?
Here is the Synth
Guys: I have been following your discussion.
Does PhotoSynth create a map that includes geocoding in it? In other words, can the overview created be imported to something like GoogleEarth so that specific areas be identified by Lat / Long coordinates? Thank you.