I'm not quite sure that they're right that this is the first autonomous transition from vertical to horizontal flight (there are some others that do that already, including Quadshot, although I suppose they could argue that it's more of a copter than a plane, or maybe not a full UAV?), but impressive all the same.
From Hackaday:
[Team ATMOS] from Tu Delft University in the Netherlands, has developed a UAV that can autonomously transition from quadcopter flight to that of a fixed-wing aircraft. Although the world has seen several successful examples of transitioning-flight or VTOL aircraft, team [ATMOS] claims to have made the first autonomous transition of this type of craft.
This UAV was featured in their school newspaper, which provides a write-up about the work that went into creating this hybrid UAV. When you’re done with that, be sure to check out the two videos after the break. The first shows the [ATMOS] taking off vertically and flying off as a flying-wing fixed aircraft. The second video shows this and other UAVs in the [DARPA] competition that it was designed for. Fast forward to 2:24 to see this aircraft do a fly-by.
Comments
@Murray I think Dave means the right project. The ATMOS team did a lot of flying in high winds as they are in the Netherlands they have plenty of that. :) They may have a video showing off their high wind performance.
I only have a video of the Quadshot dealing with some wind: https://www.youtube.com/watch?v=ZfUNnQ4dKmc
@Dave, you might be thinking about another team. For UAVForge we quickly put together a Octa-V with lifting body in the center and tilted motors. The transition from vertical to forward flight therefore required less of an angle change. It did handle winds well, but unfortunately there was that yaw problem with Arducopter 2.4( If I remember correctly) and during one test flight in high winds it kept yawing until it went out of range and crashed. Been building various other configs since. There was also a mixup in our flight demo and very few people saw the proof of flights posted on youtube. The links are here: http://youtu.be/z_ASzhw9s2g and http://youtu.be/Pgo5cLVPwt0
Still, well done to Atmos for getting it working. Especially the custom video encoding they did through cellular network.
I've been wondering about building a smaller electric version of Aerovel's flexrotor. Anyone have any ideas on this? The link is at
http://www.aerovelco.com/Flexrotor.html
I think the main problem with the design is that it doesn't do well for flights in significant wind.
@arashi not exactly they developed their controller at the end separately as they were going for full autonomity and less the funflying side of things as the Quadshot. They are actually flying fully autonomous. The whole mission on UAVForge they did was done in fully autonomous mode including takeoff.
Yes it was a bummer that one of their ESC's died in the approach for landing. :(
We are working together with the ATMOS team on integrating both the Quadshot control code and the autonomous code with the newest Paparazzi code to get both sides together.
We sponsored ATMOS with an early prototype of the Quadshot. They based their design on it and optimised it for the requirements of the competition. It is also based on Paparazzi as Quadshot is.