The team took advantage of the good weather this weekend and got a lot of flying in. We have made some great progress:

  • Fixed up navigation tuning and have got plane flying very well
  • Completed the required 5 hours of auto flight time
  • Tested long-range comms with antenna mast to 2.6km
  • Conducted several bottle drops with IV bag not bursting from 100m, and flew plane at weight of 3.5 kg

Nav Tuning

On Saturday we flew at Tompkin’s Park, situated on the Swan River. Following some good advice from the famous Andrew Tridgell of the CanberraUAV team, we adjusted our waypoint radius parameter and recalibrated the airspeed sensor. This greatly improved the navigation control loops of the UAV and meant that our waypoint tracking was very good. More info can be found here. I will be posting a blog article on APM tuning soon.

Our UAV's flight path with greatly improved waypoint tracking.
Our UAV’s flight path with greatly improved waypoint tracking.

5 hours Auto flight time

Our team has now logged a total of just over 6 hours of autonomous flight time. This meets the requirement of 5 hours logged autonomous flight for the Deliverable 3 document which must be submitted to the competition organisers by August 7.

Our UAV flying autonomously
Our UAV flying autonomously

 

Long-range comms test

On Sunday we went to Rockingham Salt Flats, which is a much larger area and suitable for our long-range comms testing. Our long-range comms is required to keep a constant connection between the ground station and the UAV for a range of up to 10km. Beyond about 500m, the UAV is no longer in Visual Line of Sight (VLOS) of the ground station – it is against the CASA regulations to operate a UAV Beyond Visual Line Of Sight (BVLOS) without special approval. Therefore, whilst Geoff controlled the UAV from the ground station, Mike went out with a walkie-talkie and RC transmitter to keep the UAV in constant visual line of sight.

Geoff at the ground control station
Geoff at the ground control station

The plane was sent to fly out further and further and then loiter in that location, while Mike walked/jogged to keep up. At 1km the comms still at 98%, but the walkie talkie dropped out of range, so we switched over to mobile phones at this point.

At 2km range, the computer did a poorly-timed automatic restart, which meant that the Ground Station was no longer controlling the UAV. Mike was able to take back control on the UAV and land it. Once the computer was restarted and the connection re-established, the plane was re-launched and the comms testing resumed.

At 2.6km, the UAV was right at the edge of lake, and the comms was still at full strength. At this point we returned home because the battery was running low, and there was no more space to go.

While we weren’t able to test the comms to the full required range of 10km, the RFD900 radio modem performed exceptionally well with no reduction in signal quality at 2.6km, which is very encouraging. Michael also got a good jog keeping up with the plane!

Bottle drops

We performed a two test bottle drops. We tested 500ml IV bags which were dropped from a height of 100m. On both occasions the IV bag didn’t burst, however there were a few slow leaks in the seams, and in one drop there were a few small punctures from the bushes.

Mike with IV bag which did not burst after being released from a height of 100m
Mike with IV bag which did not burst after being released from a height of 100m

With the IV bag attached the UAV weighs 3.5kg. A slight reduction in endurance was noted, with the plane being able to fly for about 35 minutes compared to 40 minutes previously. Our next test will be to add another battery cell to increase the endurance.

Advertisements