CanberraUAV are planning to run a UAV workshop early next year at MHV. Unlike previous workshops, this won’t be a build/fix-your-own-quadcopter workshop; rather a detailed look at some of the technologies we use in our (open-source) UAV’s.
The aim of the workshop would be to educate people on how to understand, use and extend some of the technologies we use in CanberraUAV.
We’ll conduct the workshop for both Windows and Linux users, with no previous experience or knowledge required
The specific details are:
Date: 10am-5pm Sat 4th Feb.
Venue: MHV Canberra
Topics and timings:
Autopilot (APM): 10am-12pm
Ground Station: 1pm-3pm
Stay tuned for more details!
Now that we’re all safely back home, I’d just like to give thanks to all the support CanberraUAV has been given by individuals and organisations. Without you support, we would not have made it to 1st place in the 2016 UAV Challenge.
Thanks go to:
- Gaui: For supplying us the 2x GX9 helicopters (at a steep discount) and ongoing technical support for the GX9’s over the past year.
- Halo Blades: For supplying the helicopter blades for the aforementioned GX9’s
- Hobbyking: For the store credit that allowed us to buy many small bits and pieces for our airframes
- CMAC: For letting us test fly our many experimental airframes at your field and your warmth (literally in terms of the fireplace you installed!) and friendliness to our team.
- Ardupilot: For all your awesome work in creating a world-class flight controller and depth of support for all of our issues and queries.
I’d also like the thank the entire CanberraUAV team – together have once again proven our ability to make world-leading UAV systems. Our skills run the entire gamut of software, airframes, electronics, imaging and communications. Also to those who have assisted me with the administration and organisation of CanberraUAV – thankyou as well. I deeply appreciated all the (voluntary) time you’ve devoted to this awesome hobby.
Today was a free day for us, as the UAV Challenge finished yesterday. As the original event schedule had today as a bad weather contingency day, we’d arranged to be in Dalby for the day anyway.
The morning was spent doing a few extra interviews and photos with the videographer from the BBC, who has been covering the event this week. Hopefully he’s got enough footage for a decent documentary on the UAV Challenge!
After that, we spent some time chatting with a few other teams still in Dalby and providing some advice and technical assistance with the Ardupilot/MAVProxy software.
There’s now plenty of new photos of the UAV Challenge in our Flickr account (https://www.flickr.com/groups/canberrauav/pool/), most of which were taken by Darrell Burkey.
Tomorrow we start the long road back to Canberra, and then back to our normal lives!
Today was the final flights of the UAV Challenge, then the awards and presentation.
As always, the UAV Challenge organisers have posted a good summary of the day: https://uavchallenge.org/2016/09/29/final-day-of-medical-express-2016/
From our perspective, we were content to watch and chat with the other teams.
We had the team photos and awards ceremony around mid-afternoon. CanberraUAV came 1st place (though we weren’t eligible for the $50,000 prize).
The plan from here is to relax for 1 day and start the long drive back to Canberra on Saturday.
We’ve had an awesome time a the UAV Challenge! – many thanks to CSIRO/QUT for organising the event (with assistance from their many sponsors).
It was great the meet all the other teams too; we’ve made many more friends and spent hours discussing the finer points of building and flying UAV’s. There were even a few students from UQ looking to enter a team in the next UAV Challenge.
All in all, we’ve had an awesome week!
The UAV Challenge organisers did a really good writeup of today’s events: https://uavchallenge.org/2016/09/28/day-2-of-medical-express-2016/
For my part, I’ll just add a few bits from CanberraUAV’s perspective.
We spent most the morning and early afternoon running final checks on our equipment and planning our ground station placement in the field. We also noted the wind direction and tweaked parts of the Quadporter’s takeoff/landing sequence to be into the wind.
Other than that we nervously waited and watched the other teams being called. By time the 3rd team was called up we realised the pace of the competition was higher than we thought – the chances of us flying today were looking increasingly higher.
Our flying time ended up being around 3pm-4pm. There was a light breeze, which was fine for us.
Setup for our mission went smoothly (apart of a bit a messing around with the NMEA stream to the organisers, and the camera on the Gaui not working). Takeoff of both Quadporter and Gaui went quite smoothly.
The Quadporter entered the search area first and started looking for Joe. The Gaui arrived a minute later and began it’s pre-programmed search pattern (despite the lack of camera). Soon after starting this, the Gaui reported a loss of engine power and auto-rotated to the ground (impacting the ground at approx 25m/s).
The organisers were then kind enough to let us continue the mission (noting that we were now out of the running for the full $50,000 prize). After some time searching the imagery for Joe (he was quite difficult to spot), we found him! The judges confirmed that our position of Joe was accurate and we then landed safely at the remote site.
After the blood sample was loaded, the Quadporter took off and returned to base. At this point we were running behind schedule and pushed up the Quadporter’s speed for the return journey.
We landed safely back at the base site. And that was the end of our mission!
Tomorrow has the remaining teams flying, subject to weather.
Today was a very long, very exciting day.
We started at 7:30am with team registrations and safety briefings. It was really good to meet the other teams (in many cases saying hello to familiar faces from previous competitions).
The rest of the day was spent with scrutineering (the judges checking our UAV’s are safe to fly). In total 8 teams passed. Of the original 10 teams, one 1 pulled out due to a crash a few days ago. Another (international) team still had their UAV stuck in customs and were waiting for it to be released and transported to Dalby (was delivered by the end of the day; will be scrutineered tomorrow).
Our scrutineering went well. We had one issue of a particular geofence test (with all engines running on the ground) being unable to be demonstrated, but a quick patch by Tridge fixed all that.
The afternoon was spent assisting the other teams (as required) debug their systems. I spent a few hours helping to debug an issue where the NMEA GPS stream from a few teams ground station software was not compatible with the judge’s systems. Turns out the (common) software they were using was not formatting the NMEA stream correctly. A quick message to the ground station software’s author and a fixed version was released a few hours later.
The evening was spent creating the actual mission (waypoints, etc) and running simulations.
Tomorrow begins to competition flights. CanberraUAV is 2nd last in the team ordering, so it’s likely that we’ll be flying late Wednesday or sometime Friday. Thursday is likely to be called off due to bad forecast weather.
A very busy and exhausting day for us!
This morning we split into 3 groups – one running final setup and test on our GX9 helis, another doing the same for our quadplanes. The third team travelled around likely areas we’ll be flying in the UAV Challenge, checking 3G/4G coverage and confirming our our terrain dataset heights.
Interestingly, Optus coverage is far better than Telstra’s out in the flight area.
Unfortunately, our backup quadplane had a heavy landing during a practice flight. It wasn’t heavily damaged, but is unable to be repaired to a flyable state until we get back to Canberra.
After lunch, we spent some time running some simulated flights – checking our ground station network and comms, as well as our preflight procedures.
We ran into difficulties with the ground station network box. The Raspberry Pi running the Telstra 4G dongle kept browning out and crashing. It took ~5 hours until we had a working solution of using a borrowed USB power bank to power the Pi. We still don’t completely understand why the Telstra dongle was drawing large enough amounts of power to produce a ~0.7V drop in power to the Pi. But we found a solution, which given the time pressures is good enough.
Early night for us. And an early start of 7:30am tomorrow.