• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

Wow, amazing quad video (TED talk) must see

Johan

Senior Member
#1
This is really amazing, what an incredible achievement.

Control of multiple quads using a camera system an off-board computer and on-board computer, results in super stable quads. Not really a DIY project (yet) :)

http://www.youtube.com/watch?v=w2itwFJCgFQ

Research by computer scientist Raffaello D'Andrea and his team
 

Timebom

Junior Member
#3
Very interesting and I understand most of it, but was more interested in what models of frames and setups for the long flight times. lol It also made me feel like some of the fun is gone when a computer is doing all the flying.
 

Johan

Senior Member
#4
It also made me feel like some of the fun is gone when a computer is doing all the flying.
You've got a point there, but I think there would be arial filmers out there that would love the stability. But then again, this uses camera triangulation in a controlled environment, not suitable for the open field (at least not yet...)
 

xuzme720

Dedicated foam bender
Mentor
#6
You've got a point there, but I think there would be arial filmers out there that would love the stability. But then again, this uses camera triangulation in a controlled environment, not suitable for the open field (at least not yet...)
But in an open field you can use gps. I doubt you would get the kind of precision you see here though...
 

FlyingMonkey

Stuck in Sunny FL
Staff member
Admin
#7
You guys aren't even seeing the big picture.

Imagine an area devastated by tornado, hurricane, tsunami or earthquake. Roads are gone, access is almost impossible. Trying to do search and rescue is hindered due to limited availability of helicopters. Even with that, helicopters have limitations.

Now picture being able to launch, 5, 10, 100 quad drones. Flying grid patterns, using imaging that would help locate humans by shape, body heat, breath exhalations. Quads that can enter into collapsed buildings and locate a victim, then either fly out to bring back a live rescuer, or sit there and send a signal that will bring help to the victim.

The applications are endless. This is just one idea I think about when I see these types of videos.
 

xuzme720

Dedicated foam bender
Mentor
#8
I didn't miss it. It's one of the things I bring up when people start in on banning drones. There are so many applications that could benefit from multi rotor autonomy and only an extremely limited number of drawbacks, namely spying, which they aren't even the best platform for anyway.
 

Cyberdactyl

Misfit Multirotor Monkey
#9
You guys aren't even seeing the big picture.
The applications are endless. This is just one idea I think about when I see these types of videos.
The multirotor component will explode upon society once we have another decent jump in battery technology.

As soon as an 'active' (not just a dead hover) multirotor with a modest payload can stay airborne for over 40 minutes we will see them EVERYWHERE doing an incredible amount of tasks. Multirotors simply can't transit to a location, conduct a function on station, and return, with any sort of usable radius. Or . . . be local, but be able to monitor for a relatively (~60min) long amount of time. Currently tasks are basically limited to 'local' photography. New technology that allows a healthy (~10min x 2) transit to a location, everything will change.

My feeling that is still a couple years off.
 

Johan

Senior Member
#10
Hmm, now the HK EU warehouse is in the Netherlands, that would give me some nice shipping options in the years ahead:

X express delivery by drone (add $4.99)
X parachute drop to terrace (add $2.99)*

*you get to keep the parachute :)
 

Johan

Senior Member
#11
On the serious side:

What triggered my attention in this video were indeed what FlyingMonkey mentioned: the future practical possibilities, but also the amount of control that can be achieved already in a controlled environment.

There is also the fact that when I studied Applied Physics, I did my thesis on laser triangulation range finding, which is a sort of low-level computer stereo vision, so the techniques used in this video appealed to me as well

Of course once you go outside you have wind an other external factors to deal with. But given the 'auto correction' demonstrated here after pushing the thing around maybe that is not such a problem.

Camera triangulation as used here becomes less accurate when larger distances are involved (outside) and unfortunately GPS is not accurate enough yet (well at least for civil applications). Although if you combine it with accurate altimeter / optical flow sensors, sonar maybe you can pinpoint location more accurately.

I particularly liked the part where the three quads manipulate the net.
 
#13
Being a big TED Talks fan I just happened across this vid on YouTube and although not a multirotor guy....this still warrants a big fat WOW from me. I was eager to make a new post and link here but did a search first to make sure no one else had posted already. At 500 views here on FT I figured this needed a bump to get this back at the top of the list for those who may have missed it the first go round. Amazing demonstration.
 

Johan

Senior Member
#14
Being a big TED Talks fan I just happened across this vid on YouTube and although not a multirotor guy....this still warrants a big fat WOW from me. I was eager to make a new post and link here but did a search first to make sure no one else had posted already. At 500 views here on FT I figured this needed a bump to get this back at the top of the list for those who may have missed it the first go round. Amazing demonstration.

I agree, TED is great!

Whenever I'm 'depressed' and stuck in Content Managements CSS issues, I check out TED and look at what I could have done instead... (pretentious, I know... ). :)

Somehow TED always provides a light at the end of the tunnel!

Thanks for the bump!