The battery life of quads is a bit dismal in its current state. I can typically fly for 7-12 minutes per battery, but that goes by very quickly. It also means I can cover much ground with it flying as a UAV. I’ve decided to branch out and get a glider. This will allow me to fly for 20-30 minutes on a charge, which will give me quite a bit more range. It will also fly faster in a less aggressive position, so it should cover quite an area. I’ve added the hitec telemetry sensors for the time being. Eventually I’ll move my ardupilot from my first quad into this glider.
If you’ve seen some of the near dark videos of my f450 flying, you can see how hard it is to see with a dark backdrop. I’ve started to design a lighting prototype around a TLC5940 chip and an arduino micro. I’d like to be able to do animations based on one or more inputs. We’ll see how far this goes …
I thought a gyro-stabilized gimbal would improve the video footage from my quad. I decided to go with X468 carbon fiber gimbal. The servos are driven by the naza controller. It works nicely as you can see below.
So I crashed again 
. I was at a local park and it got out of control. Rather than injure someone or something, I dropped the throttle and grounded it. Looks like I’m going to need some more arms …
After the last video I posted of the ardupilot loitering, I decided that I was trying to tune it blind. I’ve never had any other quad, and the frame on that one I had built from scratch. In order to truly know what needs to be done to improve the frame, I need to fly one that is a finished design. After much searching, I decided on a DJI F450 frame. This provided an agile quadcopter frame and I liked the Naza electronics. Unfortunately the directions are poorly written English and I put the magnetometer on incorrectly. This caused it to be out of balance and crash. After some epoxy and zip ties, she was flying properly. You can see the result in the video.
I’m slowly starting to get more comfortable with the automated modes on my ardupilot. The first step to success is loiter mode. In this mode, it uses GPS and the altimeter (or barometer) to stay in one location. You can see my first result. The PIDs do need tuning, but its not a bad first shot.
As part of my ongoing UAV development, I wanted to be able to use telemetry and have it stream video back. I’m using a 5.8 GHz immersionrc transmitter with fatshark goggles to watch the action. For the telemetry I’m using xbee pro 900 MHz transceivers. This allows me to not only get realtime feedback but also reprogram coordinates while in the air!
I decided that the the electronics enclosure vs 3 for my quad was bit crude looking. It functionally worked, but needed to be more refined. I modeled a new one with channels for the electronics. Then I lasercut a cap so that I had visibility into the case. You can see the results below. This is much more stable and refined.
For my birthday, my wife is getting me parts to build a UAV. The first said parts are in and I’m going to begin constructing the frame this weekend. Below you can see the processing and sensor units on top of the four propellers. Its much smaller than I expected!
