AeroVironment Testing (continued)

     Did more flight tests Tuesday and Wednesday. On Tuesday we did two flights in a row without checking the data. The camera settings were way off and the live preview wasn't enabled so we couldn't tell that all of the pictures would turn out underexposed. We were still having problems getting the camera settings right on day three so it will just be a trial and error process to get it right. Also, we have been using auto focus to eliminate a variable. This slows the frame rate down by about 50% but due to the high winds, the plane is flying slow on the upwind legs so we are still getting good overlap.
     We will try to get a Cannon camera running so we can write some CHDK software and hopefully be able to adjust the camera settings in flight via our telemetry stream but that is a ways off.
     Tuesday morning we had meetings from nine to one with four AV reps. They were interested in the work the Alaska Center For Unmanned Aircraft Systems Integration (ACUASI) is doing and will hopefully be a future source of support. They are also working on a high resolution camera payload like the one we are making. I think that is why they were so helpful on this trip. While our payload is not near a final stage, some of the initial testing may have been mutually beneficial. They did devote three employees to help us for three days.
     For the last flight on Tuesday, Greg decided to fly the larger lens on the camera. The lens is longer and sticks out past the buffers on the fuselage so that if the plane were to land on it, the lens would be the first thing to hit. We knew the risk and told the pilot to deep stall the plane into a bush so that the camera might be saved. After gathering data for 20 minutes and several approaches, he was able to land the plane partiality on a bush. The camera made it but the center wing took all the load and snapped.
     Wednesdays flights were fairly short. The winds had picked up again so we reviewed our photo target strategy. Since we weren't sure if we would be coming back we had pulled all the targets the night before. Instead of placing them along the road we set them out at an angle bisecting two roads in the flight area. In one flight I walked out to the targets and gave directions to keep the flight path over them. On the last flight we had the pilot do a deep stall landing on the ground so we could see where the payload would fail. I set my GoPro to the highest frame rate of 120 fps and got close to the plane hovering overhead in the strong wind. The camera's buffer fills after 15 seconds so I had to start recording after it entered the stall. When it did I took a few steps towards where it would land and held it still.
     The camera has a fish eye lens and was shooting at 720p so even though I was pretty close, the video was not very high resolution. I slowed it down and cropped the frame in Premiere; the video is below.
     After we packed up and said goodbyes at AV we had lunch and drove to LA. Greg got on an earlier flight so I waited a few hours at Steve's hotel.