After a loooong wait I have a new camera. After my initial happiness with SecurityCamera2000, I am still waiting for my replacement camera. I ordered a set of Fatshark goggles in the meantime and this has arrived in about two weeks from the order. There was a strike (yay, SA!) at some depot handling incoming parcels, and I guess my cameras are there somewhere. So, I will not blame SC2000 – yet.
Luckily, the goggles comes bundled with a receiver, transmitter and camera, so I can use that in the mean time.
I did a few FPV flights the week-end, all with me flying the place regular RC style and someone else operating the goggles.
Below is a video of the feed recorded from the goggles. Recording framerate is not indicative of what is seen in the goggles as the framerate there is fairly smooth. The video shows the typical issues I experienced – most of what seems to be fairly short bursts of interference, probably caused by polarity issues and the stock antennas.
We spent the (chilly) long week-end at the Vaaldam and I took my SPAD FPV plane with – of course.
Unfortunately, I am still waiting for my replacement FPV camera and had to make do with regular “mark 1 eyeball from the ground” RC flying. What I DID take along, however, was the el-cheapo spycam that I use to record videos in the belly of the aircraft.
I moved the camera into the wingtip this time (it’s so light that it doesn’t affect the balance of the aircraft at all) and mounted the wing on two blocks of foam.
This completely eliminated the vibration issues I had previously trying to record video whilst flying.
I took some aerial video of the place we were staying at and created image strips using Microsoft ICE (as previously blogged). I used ESRI’s ArcGIS Desktop software to ortho-rectify the photo using Microsoft’s Bing maps as reference (Bing is Microsoft’s version of Google maps, and roughly the same quality).
I was quite pleased with the result! Even though the resolution of the camera is really bad (640 x 480), it was really hard to get good stable video due to the wind and because I had to fly the plane standing outside the property to the South (making it very hard to judge the attitude of the plane) the result was better than what I expected.
Here is a comparative image, comparing to the left, my photo overlaid onto Bing versus Bing alone on the right – Notice the difference in the water level between the two images (click the images for full-resolution images).
and here is the same area, zoomed in on the North:
I’m starting to do research on the next phase after FPV whilst waiting for my camera to arrive.
There are really nice “plug and go” solutions for full autopilot and also interesting options for the tinkerer.
I will keep this post updated with links to resources as I find them.
The ArduPilot units are very good for plug and go solutions. See here.
MultiWii is a very cool and innovative project using regular Arduno boards and Wii remote controllers’ innards. See here.
Nearly had a bad “oopsie” yesterday when one of my servos became unstuck.
I felt the plane was behaving really sluggish in pitch and brought it in for a landing. It barely flared on landing and I ended up plowing through the dirt with the nose. Luckily, other than being dirty, the plane was not damaged.
Turns out one of the tail servos came loose and I was controlling the plane with only the other servo that was left.
Had this been a regular tail and it was the elevator servo that came loose, I’m not so sure I could have landed the plane intact!
That begs the question – for a future long endurance FPV platform carrying potentially thousands of rands of gear, is the redundancy given by a V-tail and twin servo aileron set up not the way to go? With that kind of set up one can lose any servo and still have control.
What are the downside to a v-tail? I can’t say that it flies much different to my other planes and it is one control surface’s drag less. So far there seem to be only up-sides…