My name is Paul Gregory and I read about this sitein the Economist article. As I now live on a small wine and olive farmin Italy, the idea of building a UAV to produce high resolution aerialphotographs of the surrounding area seems not just great fun but alsovery useful. I have been researching and reading a lot over the lastmonth and as I am now ready to start my project, it seems time to sharemy ideas and progress. Hopefully your comments will help me avoid toomany mistakes before I make them.
I have always found that a very clear goal is vital to the success of any project. My goal is:-
“Todesign, build and fly a UAV to produce a composite orthographicphoto-map of a 2km square area at a resolution of 12cm per pixel or better”
The area where I live is very hilly with a mixed terrain of forest, arableland, vineyards, olive trees and scattered farmhouses. To give you someidea of scale, the ridge on the horizon is about 2.5km away. The valleyfloor drops down by about 100m before rising up again to the ridge.There is really no flat ground around so runway takeoff is out.
There is a patch of rough ground behind the house, but this is perpendicularto the prevailing wind direction. Given the topology, I think that ahand launched sailplane with some form of air break to given asemi-stall landing is the best option. Electric trainer aircraft withwheels are out. I also think that, within reason, the bigger the betteras a larger airframe should be inherently more stable than a smallerone. Since I am learning to fly using a Multiplex EasyStar, it seemed alogical step to use the Multiplex Cularis at the basis for my UAV.
Building a UAV for phot mapping - Previous Posts
Comments
If I do not watch out all the work will be done for me! I have posed another entry about the camera and the elevation issues you cover here. I am not sure yet how to best organise my posts or how to make reference to comments in my other posts.
Regards
Paul
you have set up a clear goal which is the best way to get started.
My advices considering the image processing part as you plan to create orthos are :
- fly as high as possible (and allowed) in order to get as little images as possible to process => this means having a good camera (10Mp for example)
with a canon IXUS 900 for example flying at 500 meters elevation will give you the 12 cm you want.
At 500 m elevation, you will have about 16 ha per image with a swath of about 460 m. Depending on the shape of the area and overlap (rows and lines) you can evaluate the number of images necessary.
Of course the number of MegaPix is not the only issue for quality.
As Chris says Cularis is a good choice but you may have some issues to put the gear inside properly.
I am sure that you will get the best advices on DIY Drones to set up you system.
Thanks for the comments. I have already started thinking about total weight and COG and how to modify the foam airframe to fit the camera and autopilot. I will cover this in a later post.
Great project, with a really well-defined goal. I think the Cularis is indeed a good choice for aerial photography (the Pict'Earth guys are using one) and I've pinged them to see if it's a good choice for an autopilot, too. One of the issues with planes with the equipment compartment in the nose, as is the case with sailplanes like the Cularis, is that you're limited in how much weight you can add without throwing off the CG (or having to add lead to the tail). But given your landing terrain, I can definitely see the argument for a belly-land foamie, and if you use a light off-the-shelf autopilot like the PicoPilot and can live with a small point-and-shoot camera like a Canon SD1000, you should be ok.