Hi everyone,
I would like to know what you think is the best solution (price/quality/ease of use) for mapping around 300km (180 miles) of a highway??
We need to do photogrammetry of this project and I have suggested the use of a UAV instead of a manned aircraft.
I imagine that we will need a fix wing solution (or solutions) for doing this project but I would like to have further information.
Any information is useful.
Thanks!!
Replies
My vote for diy affordable mapping is a platform based on the Skyhunter. Enough room for NEX style camera with good vibration reduction and brushless gimbal and room for a couple big batteries (2 x 4000 mah). The fpv folks use this plane for some major distance.
I'm wondering what your resolution and accuracy requirements are? If you require a high accuracy "photogrammetric" product, then that will require a lot (?) of well spaced DGPS ground control points to obtain good triangulation results. If you only need accuracy of a few meters throw up a Skyhunter or SkyWalker with a pixhawk and go at it.
Do you need 5cm GSD or 1m GSD?
I'm hoping to be doing some road map testing with my Skyhunter in the next couple months. What I'm wondering about is how many GCPs I need to maintain high accuracy along a 50km road segment. I'm planning on staying semi-legal by driving behind the drone and keeping it in sight.
Hi Peter,
Thanks for your reply.
We are looking into an accuracy of 1m GSD. I think I will put together a Skyhunter or SkyWalker and give it a try. We do have a DGPS and will take note on the control points.
Please let us know how your road map testing comes out.
Hi,
If you require 1m accuracy on your contours and detail you might as well get onto Google maps pro and get your DEM and detail from there.
The density of your ground control points depends on your altitude of flight and camera. The points need to be placed on the overlaps of your strips. There is quite a bit of planning and prep involved in getting optimal use from ground control to manage the accuracy of your job. The Mission Planner's auto WP routine works extremely well in planning these with the pic dimensions and speed considered. I achieve < 30mm accuracy with my quad at 50m relative altitude and 5m/s speed.
IMHO, for a project like this stick with the pro's who design and make mapping, GPS and survey solutions. The guys from Trimble have a brilliant mapping UAV mapping solution that has matured quite a bit over the past years. It has been around and would serve you well. Check www.trimble.com. It costs, but you pay for a product that have done the full R&D cycle...twice and have been in the marketplace quite a few years.
Your project sounds very exciting. Good luck and keep us posted.
Antonie
My vote is for the Skyhunter. I also have a Slywalker, but there is very little room inside that fuselage compared to the Skyhunter. The Skyhunter has huge space, enough that a serious gimbal system can be incorporated. CHEAP. CHEAP. CHEAP. Bought mine from FPVModel
If one could keep a camera at nadir, to a 1/2 degree or so (using a brushless gimbal setup), horizontal offset errors could be kept to under 1 meter. So assuming (ha ha) autonomous gps accuracy of 3-4m, it might be possible to create a product with overall horizontal accuracy of 5 meters with no DGPS control (just tie points). Or depending on what software you are using, a few good DGPS ground control points might be sufficient to obtain 1-2 pixel accuracy (via triangulation). But 300 km is a long road!
A much easier solution, because you only require 1m GSD, is to order a narrow corridor of .5m satellite data (WorldView-2, GeoEye-1 or Pleiades). In my work I use these satellites a lot. Check the archives of these companies. Lets see 300km x 3km = 900 km2. What for a sale get the data for around $15 km2 (maybe cheaper). So 900km2 x $15 km2 = $13,500. Processing only requires a few GCPs. The beauty of using satellites is that the entire scene is acquired in a couple seconds, so colors, shadows, etc remain consistent across the scene. And you are only processing one image instead of 100's. Might be the right tool depending on your budget and time frame. I've mapped many roads this way. Just my 2 cents. But drones are more fun!
Thanks Xavier!
Hi Silverzorra,
I will just share my experience here. We have done a project of mapping a 65 sq. km area within a week. The straight line flight was more than 500km. The platform used was skywalker with custom arrangements. If its possible in your location to fly out of sight, you can complete the project within 4 days of flying. Even less if everything is perfect.
Regards,
Pravin
Hi Pravin,
Thanks for the insight. Yes, we can fly out of sight. Did you put together the skywalker yourself or it was a RTF solution from a third party?
Regards
Hi Silver,
I'm guessing you don't want to spend the money to get into existing "serious" $30,000.00 + platforms to do this.
For what you want to do there will be some obstacles.
First are you going to want to and can you legally fly beyond normal RC radio range?
That is problematic but not impossible.
Are you planning on full flight continuous FPV and telemetry?
That may also require extended range equipment.
And keep in mind any intervening terrain between you and the plane will interfere with all radio signals in or out (and is generally something you would like to avoid).
For the length of flights you are proposing, electric may be more limiting than you might want and you might want to consider a 2 stroke or better a 4 stroke gasoline engine, but that also introduces it's own problems.
Please take a look at my dronesarefun.com site Build an Airplane page (especially the last two planes) for an idea of what airframe might actually work with sufficient capacity and longevity for being a useful tool for you.
http://dronesarefun.com/BuildAnAirplane.html
The first of the 2 isn't quite available yet, but should be soon.
The Pixhawk is probably the best affordable autopilot for this kind of venture you can currently buy, but get a genuine 3DR one or a VRBrain, the last thing you need is quality problems with the autopilot.
This high end hobby level stuff is just beginning to enter the level of work capability you are talking about, but with careful preparation and rigorous learning, design and shakedown, it is at least now becoming possible.
Best of Luck,
Gary
Hi Gary,
Thanks for your reply. Here are some answers:
-Yes, we can and want to fly beyond normal RC radio range.
-Yes, we would like to have continuous FPV and telemetry but is not a deal breaker.
I have been checking your site and has been extremely helpful.
check out our newest Aeromapper 300, 1.5 hours endurance. You will need svereal flights but it is very reliable, many payloads supported, easy to use, handlaunchable, long range data link and control, and affordable:
http://aeromao.com/aeromapper_300
despite is has been flying already for many months haven't been officially annouced, which will happen very soon!
-
1
-
2
-
3
-
4
of 4 Next