I've been operating a Pixhawk piloted aircraft and quad for about a year. Recently I bought a DJI Phantom and was impressed with how easy it was to work with the pictures that it took. The pictures aren't as high quality as I need though. Right now I have a Sony a5100 and it's a real bear getting GPS information attached to the photos from the dataflash logs. Constant issues with missing pictures or unmatched logs. I've just about finished writing my own Geotagging program to run post flight with the dataflash log and pictures taken during the flight. This requires being really well organized when doing many flights in a day. It would be so much smoother if this was done before the aircraft landed. My question is this:
What would be the best way to geotag photos in flight? I'm thinking that I need some combination of the following things:
- A computer on board the aircraft.
~ Something like the Navio and Raspbery Pi, or Erle would be a clean solution, remove the Pixhawk entirely. However I'm not sure how smooth the interaction is between the computer and sensors. I'm guessing that a program that handles the different parts will need to be on at least two threads (triggering, copy/pasting, adding Exif info). The camera still needs to be able to trigger at less than a second (if possible!). I know the camera can do it, but maybe not with all these other things going on.
~ Or something like the Raspberry Pi Zero, or Edison, that interfaces with the Pixhawk.
- An interface between the Camera and computer.
~ The computer needs to copy the pictures off the camera as soon as they are taken. I'm not sure if this kind of simultaneous read/write is even possible on a camera's SD card.
~ A wifi module? I know the Sony camera has NFC sharing, but I'm not sure if this is suitable
- An interface between the Autopilot and computer.
~ I would want to add the GPS, barometer and attitude information into the picture.
Writing the data in won't be much of a problem. Maybe just a file that matches the picture name, with the Exif data so that as soon as the mission is over and the triggering has stopped, it can start writing the data into the pictures. Maybe this is the simplest solution.
I'm hoping some people can point me in the right direction so I'm not buying and building needlessly. Thanks,