Welcome to the Time Trust Trial contest, Round 4 (T3-4)! This round is an aerial imaging task. Here is your assignment:
Program your UAV to take photos from an altitude of ~400 feet that you then stitch together to make a single image showing an area of 500mx500m (a quarter square kilometer). Somewhere in that image, a Santa (or replica of a Santa, poster of a Santa, just you wearing a Santa hat, etc) must be seen. NO PHOTOSHOP (ie, the sample above would be disqualified)--you actually need to bring something Santa-ish to the field so your UAV can capture it in its shots (yes, I know it will be very small. Just circle the location in your image so we can enlarge and inspect--don't make us do "Where's Wally"!) This is to prove that the shots aren't actually taken from Google Earth ;-)
You can use any path strategy you want: "lawnmower", spiral, concentric circles, etc...
For stitching software, I use the free PTGui, but you're welcome to use whatever software you prefer.
And for your camera, may I suggest you hack up some cool way to trigger the shutter with our cool ServoSwitch?
KML tracks must be provided. Video is not required, but is suggested.
We've now switched to a six-week cycle, so the deadline is 12:00 midnight PST on Sunday, January 17, 2010 now Monday, January 18th due to the Martin Luther King holiday in the US.
The overall winner will be the best quality image, as decided by the judges (based on a combination of resolution, stitch quality and overall coolness--a clever Santa will win you brownie points, and a pretty area is no bad thing [note: snow is lovely, but be warned that it can confuse stitching software]), but everyone who completes the challenge will win a prize.
Enjoy!
This software is available for the express permission of non-corporate, single-user environments and should be considered a Beta product and treated as such. We reserve the right to remove the download link at any time.I make no guarantees on the usability of this software, and all credit for the development of this software goes to JB Scoggins. It should work under windows xp 32 bit. All other windows flavors are not guaranteed to be usable, however, it should be ok under windows 7 32 bit.
It would be greatly appreciated if you could forward feedback and bugs to samc99us AT yahoo DOT com.
I'm a little baffled by the scoring too...if you look at the post count to win ratio, I would say the scoring is crap!
That aside, I still think this was a great competition. For the most part, this is very similar to the AUVSI Stundent UAS competition held annually in MD. That is a world renowned competition, and it's a testament to this site that a similar competition could be organized online.
Couple of thoughts:
1) We don't get snow this far south, so that was not an option and imo should not have been part of the scoring
2) We are not limited to weight in the camera. We built the plane around the camera because we've done this for many years and know that is the limiting factor. We are running a Nikon D60 with a 18 mm VR lens set at infinite zoom. The photos shown here (namely, the high res ones you need to load into google earth) show that even at 300' the resolution and photos aren't sharp. For reference, a D50 has worked better for us in the past than this new D60. May be looking at a D90 for an upgrade if we cannot get the sharpness we want.
3) For those doing auto-landings, flying lower and slower is a good way to improve resolution. We are dropping down to 200' for our next imagery run to get better resolution. We have software that builds our flight paths and computes our overlap. This is available to all for free on ncsu aerial robotics wiki (down atm). Of course, the problem is getting a wider angle lens on a point and shoot.
4) We have the software to do orthorectification. It is slow, on top of the line machines (Core i5+8gb ram, macbook pro's etc.), especially with the high MP images everyone is taking. The easiest, cheapest solution for us was to build a two axis gimbal and drive it with the IMU in the autopilot. This gets us to within plus minus 5 degrees for pitch and roll. Thus, we try and take the important photos when the plane is flying straight and level.
Once again, thanks for the competition and I'm glad to see stiff competition from the smaller teams.
Yes this round was the first real task, and we will revert to silly tasks next!!!
As for orthorectification, we were honoured to have an entry from one company that is really doing that!! I have had my eye on Smartplanes for a while now, and blow me down they enter!
I hear your comments about altitude Chris, but realistically thats what we are going to have to put up with, so best we make it work at 400'
Its hard to sum up well what you have all achieved, and to some extent make look easy.
Just travel back in time three or four years and we were all talking about keeping an airframe level and turning towards way points, then making the whole thing lift something of high enough resolution for long enough to be useful.
Now the worry is at what altitude and how to get accurate shots.
For all those that did not enter because obviously its too easy, have a crack yourselves, every entrant displayed an understanding of many facets of UAS flight, all of them needing to come together at the same time to make the images happen.
Right everyone go out and buy a big motor and small battery for the next round.
Congratulations everybody,
I would like to hear a summary from all of you about the contest.
My observations are:
-stitching of the photos below 150-180m is uncomfortable because of low shutter speed
-we are limited to around 300g camera weight on all platforms
-photos over snow reflect a lot of sunlight allowing sharp photos even from cheap cameras
-we have tested a few landing patterns: naive approach with directly drawn legs like on dot matrix printer
-as above but with additional outside loops, similar to what cropcam does, guaranteeing straight passes in the middle by moving turning overhead to the outside
-convoluted scans, probably most efficient for area vs travel length (brakar, Steve Joyce)
-stitching the photos in hilly area can be very difficult, also because of parallax errors of then-close objects like trees. there, 200m AGL sounds like minimum
I am really happy of the contest because we finally managed to 'do something useful' with our UAVs. This is literally opening our hobby for the 'outside world': we do not have to drop a bomb or two in order to be noticeable. I regret the absence of Cropcam.
Ok I have had more than one PM for the result, so without Steves airframe........ Remembering everybody gets a prize this time!
Brakar is commended for trying so hard and completing a round without landing in a tree, Brian demonstrated why he needed that disguise from earlier, living near the planet of the Apes as he does, everybody did a top job!!
Now the overall standings after four rounds:-
Vassilis 24
Brakar 23
Brian Wolfe 21
Mark Griffin 18
Krzysztof Bosak 17
Andrus Kangro 12
Jesse Jared 8
IOS 6
Bill Premerlani 6
MarcS 6
Joe 6
Steve Joyce 5
Steve Westerfield 3
Chris Anderson 3
Icebear 1
Well done again everybody, I think the next round will appeal to a larger portion of the crowd Chris will announce it soon! If you have not already guessed.
Right just awaiting absolute confirmation of autopilot for Steve Joyce and airframe from Steve Westerfield and then I will post!! Must say this round has been the hardest so far, everybody has done a great job and lots of photo stitching issues have been spoken of, very cool. Get those PMs back to me chaps.
Comments
Ok, does someone has an example of a way to trigger the shutter with the mentioned servoswitch?
Our router was down until a little while ago, hence no access.
http://art1.mae.ncsu.edu/twiki/bin/view/Main/InstallPathPlanner
This software is available for the express permission of non-corporate, single-user environments and should be considered a Beta product and treated as such. We reserve the right to remove the download link at any time.I make no guarantees on the usability of this software, and all credit for the development of this software goes to JB Scoggins. It should work under windows xp 32 bit. All other windows flavors are not guaranteed to be usable, however, it should be ok under windows 7 32 bit.
It would be greatly appreciated if you could forward feedback and bugs to samc99us AT yahoo DOT com.
That aside, I still think this was a great competition. For the most part, this is very similar to the AUVSI Stundent UAS competition held annually in MD. That is a world renowned competition, and it's a testament to this site that a similar competition could be organized online.
Couple of thoughts:
1) We don't get snow this far south, so that was not an option and imo should not have been part of the scoring
2) We are not limited to weight in the camera. We built the plane around the camera because we've done this for many years and know that is the limiting factor. We are running a Nikon D60 with a 18 mm VR lens set at infinite zoom. The photos shown here (namely, the high res ones you need to load into google earth) show that even at 300' the resolution and photos aren't sharp. For reference, a D50 has worked better for us in the past than this new D60. May be looking at a D90 for an upgrade if we cannot get the sharpness we want.
3) For those doing auto-landings, flying lower and slower is a good way to improve resolution. We are dropping down to 200' for our next imagery run to get better resolution. We have software that builds our flight paths and computes our overlap. This is available to all for free on ncsu aerial robotics wiki (down atm). Of course, the problem is getting a wider angle lens on a point and shoot.
4) We have the software to do orthorectification. It is slow, on top of the line machines (Core i5+8gb ram, macbook pro's etc.), especially with the high MP images everyone is taking. The easiest, cheapest solution for us was to build a two axis gimbal and drive it with the IMU in the autopilot. This gets us to within plus minus 5 degrees for pitch and roll. Thus, we try and take the important photos when the plane is flying straight and level.
Once again, thanks for the competition and I'm glad to see stiff competition from the smaller teams.
As for orthorectification, we were honoured to have an entry from one company that is really doing that!! I have had my eye on Smartplanes for a while now, and blow me down they enter!
I hear your comments about altitude Chris, but realistically thats what we are going to have to put up with, so best we make it work at 400'
Its hard to sum up well what you have all achieved, and to some extent make look easy.
Just travel back in time three or four years and we were all talking about keeping an airframe level and turning towards way points, then making the whole thing lift something of high enough resolution for long enough to be useful.
Now the worry is at what altitude and how to get accurate shots.
For all those that did not enter because obviously its too easy, have a crack yourselves, every entrant displayed an understanding of many facets of UAS flight, all of them needing to come together at the same time to make the images happen.
Right everyone go out and buy a big motor and small battery for the next round.
I would like to hear a summary from all of you about the contest.
My observations are:
-stitching of the photos below 150-180m is uncomfortable because of low shutter speed
-we are limited to around 300g camera weight on all platforms
-photos over snow reflect a lot of sunlight allowing sharp photos even from cheap cameras
-we have tested a few landing patterns: naive approach with directly drawn legs like on dot matrix printer
-as above but with additional outside loops, similar to what cropcam does, guaranteeing straight passes in the middle by moving turning overhead to the outside
-convoluted scans, probably most efficient for area vs travel length (brakar, Steve Joyce)
-stitching the photos in hilly area can be very difficult, also because of parallax errors of then-close objects like trees. there, 200m AGL sounds like minimum
I am really happy of the contest because we finally managed to 'do something useful' with our UAVs. This is literally opening our hobby for the 'outside world': we do not have to drop a bomb or two in order to be noticeable. I regret the absence of Cropcam.
Brakar is commended for trying so hard and completing a round without landing in a tree, Brian demonstrated why he needed that disguise from earlier, living near the planet of the Apes as he does, everybody did a top job!!
Now the overall standings after four rounds:-
Vassilis 24
Brakar 23
Brian Wolfe 21
Mark Griffin 18
Krzysztof Bosak 17
Andrus Kangro 12
Jesse Jared 8
IOS 6
Bill Premerlani 6
MarcS 6
Joe 6
Steve Joyce 5
Steve Westerfield 3
Chris Anderson 3
Icebear 1
Well done again everybody, I think the next round will appeal to a larger portion of the crowd Chris will announce it soon! If you have not already guessed.
-
1
-
2
-
3
-
4
-
5
of 7 Next