In the last article I described how the MicroWing UAV was constructed. Finally the time has come to head off to the flying field with a lot of cameras…
MicroWing UAV Recap…
The MicroWing UAV is a small, fully autonomous aircraft designed and constructed using foam board building techniques. Born from a combination of the VersaDrone and MicroWing projects, the aim was to build a low-cost platform that was able to perform tasks similar to commercially available RPAS costing tens of thousands of dollars.
Back up a bit…
Last time I forgot to mention one other innovation used in the construction of the MicroWing UAV. On previous KFm wings I had simply wrapped the leading edges in cloth tape. The strength in the tape itself giving form to the leading edge of the wing. For the MicroWing UAV this simply wasn’t going to be good enough and it was my fellow flyer Andrew who suggested the solution. On his KFm6 “Sausage Tray” wing he finished the leading edge steps with fillets of Depron foam. Depron is a lot easier to sand and shape than foam board and it was very easy to build a smooth leading edge.
On to the flying then!
As is usual with the weather around here the conditions were less than perfect when I headed off to the LMMAC field for test flights. It was grey and blustery with the threat of rain in the air but I simply couldn’t wait any longer.
First flights were conducted under full manual control and the result was something of an anticlimax – the MicroWing UAV flew perfectly!
The KFm7 section performs very well in the air generating more lift and with better slow flight handling characteristics than the symmetrical sections.
You have to love the versatility of the Mobius HD camera! Over the course of a couple of batteries I managed to get plenty of good footage. Of course, the MicroWing UAV isn’t really for shooting cool airborne footage. I have yet to mount any kind of downward facing stills camera so that might change things around a bit.
With the 1000mah 3S packs I was getting around 8 minutes of flight time. That is a little less than I had hoped for but it is a very small pack and a larger motor than the original MicroWing.
As a final test for the day I decided to make some flights in Autotune on the APM to try to get those PIDs locked in. That’s when things went wrong…
Autotune in Ardupilot places the aircraft into FBWA (Fly-By-Wire). You have to push the sticks all the way over and hold for two seconds for the APM to “learn” the extremes of movement and then tune itself accordingly. This just wasn’t happening with the MicroWing UAV. No matter how much I threw the sticks around the airframe barely responded.
The problem, as it turns out, is the OrangeRx software. When you switch on one of the wing mixes in the menu it severely restricts the servo movement to account for the mixing. This means that, as far as the APM is concerned, the PWMs never reach the required extremes to register a response.
Time for some modifications
Back in the workshop I had some decisions to make. I really wanted to use the OrangeRx, just to prove it could be done but things just weren’t going to progress quickly enough. I decided to remove the OrangeRx unit and replace with a Taranis receiver. Using the Taranis would give me much better control over the settings and flight modes anyway.
…and back to the field!
So, the Taranis gave me the required PWM outputs to make the Autotune process work. Things went pretty well and I soon had the MicroWing UAV flying around in Stabilize and FBWA without a problem.
A quick test of the Return to Launch function showed us that the navigation was working fine so it was time to plug in some waypoints and see how it flew under full autonomous control…
…not so good!
Basically, whilst it was tracking to waypoints okay, I kept losing altitude particularly after the turn. I had put this down to a speed issue but a similar problem recently with a multirotor has highlighted the fact that I probably had absolute altitude checked in mission planner and this can cause just such a problem.
Who knows? We are now up to the present day so I need to get back to the field. Check the tuning and try some more autonomous flights with the altitude corrected. After that it’s just a case of fitting the payload.
Watch this space!