It's time for the second round of the second season of the Trust Time Trial (T3) competion! This round is an autotakeoff and landing round, with 3D waypoints. Like the last round, there are two divisions: plane and copters.
Your mission in this round is to complete the course diagrammed above. Your UAV must autonomously do the following:
- Autotakeoff. NO MANUAL CONTROL. This must be fully autonomous launch, with the RC transmitter on the ground. For aircraft we will accept rise-off-ground, catapult or hand-launch (but remember, no transmitter!). For copters, it must rise-off-ground. Just flick your RC transmitter's mode toggle switch and put that bad boy on the ground while you watch the show.
- Hit the above eight waypoints, which are in increasing and then decreasing height (the target altitude is given after each waypoint number). They are all 100m apart, for both planes and copters. The exact position of WP8 (or however more waypoints you need to end the mission) is not important--it is just whatever you need to set up your landing pattern properly.
- Autoland, as close to your launch position as possible. Again, no transmitters!
To enter, post the following in the comments below:
- KML track and screenshot of your track in Google Earth, with launch and final rest positions marked.
- A description of your vehicle and autopilot.
- Because this is a hands-free round, we want videos of the whole thing! Upload them to YouTube and embed the video in your entry.
Gary Mortimer, as always, will be the judge. In this case it will be scored primarily on "style": he will use qualitative rather than purely quantitative measures to reward clean patterns, tight altitude control and the closeness of the landings to the takeoff points. Note: experience has shown he can be influenced by creativity and fun, so make those videos count!
There will be three winners in each category (planes and copters), with the #1s getting a $100 gift certificate at the DIY Drones store and #2 and #3 getting $25 each.
All entrants who successfully compete the course with a copter, regardless of place in the judging, will receive a flying robot merit badge!
This is a pretty tough challenge and the weather is still dicey in the Northern Hemisphere, so you get two months for this one. Deadline is Sunday, March 4th, at midnight PST.
Comments
Ohhhhh a Twinstar, plus 5 points I have a soft spot for them, also a tight looking launchsite. Well done.
Well done Colin
So the only real UAS drivers around here drive aircraft ;-)
Just noticed your from Bristol, I was born there and parents are still in South Somerset, so you might be building a lead. All those positive points might be deducted if Lulsgate noticed. I have a story about a manned flight exactly where you are flying, but thats for the pub not here!
My post was too long, here's the rest of it:
My airframe is a Multiplex Twinstar II. I’ve modified it with brushless outrunners, the details can be found on my blog. The autopilot is an APM 1 with mediatek GPS (sadly I broke my ublox trying to get a tricopter into the competition). I also have a 5883L mag, but being the first outing with both it and the mediatek, I’m not sure how much difference it made.
I was running the APM 2.8 code, but removed the two lines that scale the navigation roll term with ground speed.
I’d like to thank all the dev team for sharing their hard work; Chris and Gary for these competitions, please keep them going, I’m learning a lot from taking part. Also, thanks to my club-mate Paul for helping film this round.
And finally, a gentle reminder of the peril at my flying field: here’s where my plane “landed” once back in the pre-beta days
Hi all,
I’ve now flown my entry for this round. I did the course two times in a row to test the consistency and prove that the landings weren’t too hard ;-)
The flight times were almost identical at 61 and 62 seconds each. The landing spots from the logs were measured to be 7 metres apart in google earth. The KML for the flight can be found here
Here’s a YouTube video of the two flights
If you look carefully, you can see my arm go out away from the transmitter (which I had to hold to flick into auto mode) after launching. Also, I held the transmitter up during the landing to show that I wasn’t flying it.
The pitch up just before landing is where I scheduled a speed reduction. The weight of the on-board camera caused the canopy to pop off on landing, a bit of tape would have helped...
Here’s a top down view of the flight paths. The red line is the first flight and is the log I am submitting for the competition. The biggest deviation seems to be from my launch direction :)
The big jaggies in the red must be data errors because the flight itself was smooth. The green flight actually lost GPS fix twice, once straight after launch and once about 20 seconds in, for half a second.
Here’s a couple of 3D views
and the measurement between the two landing spots
I measured my climb rate response using the FBW-B mode. There’s a bit of a lag in the step response, which I thought could be an issue on such a tight course, but the results came out pretty well I thought :)
My airframe is a Multiplex Twinstar II. I’ve modified it with brushless outrunners, the details can be found on my blog. The autopilot is an A
Points may be deducted for people that wear cameras on their hats though. That's just silly.
Ahh yes Ric. Good Point, Points for Peril ....
Deliberate cross wind take off and landing with wind blowing from left to right !
Launch plane angled with right bank to make plane have to correct for launchers ineptitude !
Launch with trees just to the right of the flight path ... just to add some additional entertainement. :-)
Strikes me that the UAVDB gang are the ones firmly at the cutting edge of things and willing to risk their gear to try stuff out. Pie and pints on me all around if I ever catch up with you. Only one per entrant, terms and conditions apply.
@igal wheels are a drag at our size of airframes.
The record breaking Zephyr (14 days in the air, made with real ale and chips) is hand launched because any one of the team can detect an issue with the airframe and call a halt and if the field is big enough they can always turn and run into a changing wind.
@Robert any one can enter, even Chris ;-)
@Phil, I think extra points are also available for people that have checklists on their airframes. That makes lots of sense and should be rewarded!
Quote
Points are always available for peril!
I think you earned some, watching the on board video and seeing how close your plane went to the trees, isn't it Gary ?
Best regards,
Ric
Hi Pete,
congratulations for a great entry !
Best regards,
Ric
Hi Folks, well I suppose its time to keep this competition moving along with more entries ....
The plane is an Alpha Axion 135 Brushless flying a UDB4 with MatrixPilot from trunk (revision 1262).
Ublox GPS running at 2Hz, and a magnetometer from the DIY Drones store embedded out in the left wing.
Flight Video taken from my hat is here.
Flight VIdeo taken from on board the plane is here.
(Further screencast of flight analysis of the flight here.)
Here is the Google Earth KMZ file.
Thanks to DIY Drones community for putting on this event. And Thanks as always to Bill premerlani and all the members of the UAV DevBoard community for their advice and help.
Best wishes, Pete
Hand lanching is for the weak arduinos .
Real man taking off from ground .
-
1
-
2
-
3
-
4
-
5
of 5 Next