Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Intelligent Autonomous Flying Robots Learn and Map Environment As They Fly

timothy posted about a month ago | from the learning-as-they-go dept.

AI 37

An anonymous reader writes with this story about a machine-learning project out of the UK's University of Sheffield: Using simple drones the researchers have created automatic-control software that enables the "flying robot" to learn about its surroundings using a camera and an array of sensors. The robot starts with no information about its environment and the objects within it. But by overlaying different frames from the camera and selecting key reference points within the scene, it builds up a 3D map of the world around it. Other sensors pick up barometric and ultrasonic data, which give the robot additional clues about its environment. All this information is fed into autopilot software to allow the robot to navigate safely, but also to learn about the objects nearby and navigate to specific items.

cancel ×

37 comments

I for one welcome (-1)

Anonymous Coward | about a month ago | (#47343411)

whatever. I give up.

Fly My Pretties (-1, Troll)

Anonymous Coward | about a month ago | (#47343421)

Seek the brown skinned raghead terrorists. Seek and Destroy!!!!

Love,
Every American

Re:Fly My Pretties (-1)

Anonymous Coward | about a month ago | (#47343423)

EveryAmerican is a psychopath. JFYI.

Re:Fly My Pretties (-1)

Anonymous Coward | about a month ago | (#47343453)

America went crazy after 9/11 and never recovered.

Pretty Important given the rapture and all (0, Troll)

Anonymous Coward | about a month ago | (#47343463)

This is pretty important given the biblical version of the end times. A hole in the ground opens up and from the underground comes small golden flying things to seek out man and sting him. Sounds like small drones to me, but i thought not in my lifetime as they need an operator to control them and software for self flying self control self knowledge doesnt exits and likely wont in my lifetime, but here we have self learning self flying self targeting activated software. There goes my rational for the rapture not being in my lifetime. Geez and to think the end may come in my lifetime, so fast from zero to total control by the beast power.

Re:Pretty Important given the rapture and all (0)

Anonymous Coward | about a month ago | (#47343477)

This is pretty important given the biblical version of the end times. A hole in the ground opens up and from the underground comes small golden flying things to seek out man and sting him. Sounds like small drones to me, but i thought not in my lifetime as they need an operator to control them and software for self flying self control self knowledge doesnt exits and likely wont in my lifetime, but here we have self learning self flying self targeting activated software. There goes my rational for the rapture not being in my lifetime. Geez and to think the end may come in my lifetime, so fast from zero to total control by the beast power.

That's crazy. The rapture is a myth. Christians will remain to be tested during the tribulation and won't be magically taken away.

Re:Pretty Important given the rapture and all (1)

mark_reh (2015546) | about a month ago | (#47344113)

I want some of whatever it is you're smoking!

Re:Pretty Important given the rapture and all (1)

Jason Goatcher (3498937) | about a month ago | (#47346389)

They're smoking Jesus, the best high of them all.

Applications (1)

Anonymous Coward | about a month ago | (#47343469)

It would be nice if these were fed back in real time to a remote monitor. Maybe a 21st century canary in a coal mine? Applications for search and rescue, scouting real time optimal traffic routes for police / fire / paramedics.

Re:Applications (0)

Anonymous Coward | about a month ago | (#47343491)

It would be nice if these were fed back in real time to a remote monitor. Maybe a 21st century canary in a coal mine? Applications for search and rescue, scouting real time optimal traffic routes for police / fire / paramedics.

Or maybe spying on the humans whilst they plot world domination? Ever think of that?

Re:Applications (0)

Anonymous Coward | about three weeks ago | (#47357391)

"If you choose to decline further service, we may consider you in default of your contract, and your fee may be suspended, pending litigation. Good luck, Material Defender. Dravis out."

Does anybody know more details about this project? (1)

catchblue22 (1004569) | about a month ago | (#47343523)

Cool project, but the article/video is short on detail. I'd like to know more about the way this robot is actually learning. Is it a neural network? How does it know an oscilloscope is an oscilloscope? Does it use binocular vision to recognize distance? Ultrasound? Both? What type of computing hardware is on board? For that matter, what type of quadracoptor is this? And more importantly where can I get one?

Re:Does anybody know more details about this proje (1)

Anonymous Coward | about a month ago | (#47343639)

So the robot starts with no information about its environment and the objects within it. By overlaying different frames from the camera and selecting key reference points within the scene, it builds up the 3D map of the world around it. Barometric and ultrasonic sensors give the robot additional clues about its environment. All this information is fed into autopilot software to allow the robot to navigate safely, but also to learn about the objects nearby and navigate to specific items.

Instead of a neural network, the researchers used just basic game theory to program the quadrocopters. In this framework, each robot is a player in the game and must complete its given task in order to "win" the game. If the robots "play the game" repeatedly they start to learn each other’s behavior. They can then perform their task successfully – in this case getting past the other robot – by using previous experience to estimate the behavior of the other robot.

As to your question about where to get one of these, this is just a research project. It would require a lot of more hard work, some Asian manufacturing arrangements, a global supply chain, and an investor to make this a product. Maybe some day.

Re: Does anybody know more details about this proj (0)

Anonymous Coward | about a month ago | (#47345153)

Welcome to 1990, where we can share a project like this on a thing called the 'internet'.

Re: Does anybody know more details about this proj (0)

Anonymous Coward | about a month ago | (#47345179)

Welcome to 2000. We can share the software on this thing called the 'internet'. Quad copters are a dime a dozen.

Different than the SLAMM stuff at Carnegie Lab (1, Interesting)

Anonymous Coward | about a month ago | (#47343525)

Carnegie Mellon folks developed the SLAMM algorithm (and variants of if) some years back to do live mapping on their quadrotors. It has been used by almost everybody who is doing autonomous flying robots. It is hard not to say that anyone was not influenced by that work. Some of their work had laser scanners that would map the surroundings and identify walls -- building out a maze of sorts as it explored. Heck on seeedstudio.com you can pickup a (LIDAR) 360 2D laser scanner and algorithm to build your own.

Not taking anything away from the English (as what they have done is pretty impressive and a different approach), though am curious how it is different algorithm wise?

One of the research avenues I think at some point will be to have mesh networking on these robots to share info so that in a hostile/harsh environment as these things get damaged the mapping data is passed along for fire crews, soldiers, HAZMAT, etc.

Re:Different than the SLAMM stuff at Carnegie Lab (2)

K. S. Kyosuke (729550) | about a month ago | (#47343899)

Lidars are expensive, stereo cameras are cheap. Give it a good eye separation and you're set.

Re:Different than the SLAMM stuff at Carnegie Lab (1)

citizenr (871508) | about a month ago | (#47344153)

they used PTAM and replicated Jens Nyman's thesis from 2 years ago

https://github.com/nymanjens/a... [github.com]

Not impressed. (0)

Anonymous Coward | about a month ago | (#47343541)

Eh, it's just a roomba in three dimensions.

Re:Not impressed. (2)

jones_supa (887896) | about a month ago | (#47343653)

Indeed, who would be interested in an vacuum cleaner capable of smoothly navigating in 3D space. At least design something practical, like a TV integrated on my fridge.

SLAM? (4, Informative)

Animats (122034) | about a month ago | (#47343543)

Doing this is called Simultaneous Localization and Mapping, or SLAM. There's been enormous progress in that in the last decade. The basic idea is to take a large number of images of the same scene, possibly with inacccurate data about where they were taken, and build up a 3D model. It sort of works most of the time. Some algorithms do well indoors, especially where there are lots of strong edges and corners. Those are easy features to lock onto. Outdoors is tougher, although outdoors you can usually use GPS. It's a basic capabiilty robots need.

The video is frustrating. There's no comparison with previous work. Is this an advance, or did they just use known algorithms. [openslam.org]

Re:SLAM? (1)

Anonymous Coward | about a month ago | (#47343813)

If I can expand on what you said: specifically using a large collection of photographs to generate a 3D model is known as "Structure From Motion"(SFM) or Photogrammetry. It's a popular thing to do with aerial photos taken by drone/UAS for GIS data gathering and there were a number of applications for that purpose, although many of the free online ones have been commercialized last I went looking. Catch123 is one such product from Autodesk, although if you want more control over the pipeline then doing it yourself is the way to go. This comes with the caveat that SFM requires significant computing power(which touches on why weight-constrained aerial vehicles normally use a terrestrial backend for this type of heavy lifting). The fact that mobile devices in your pocket are approaching a point where this will no longer be necessary is evidenced by Google's Project Tango and probably this article. Other software products available include VisualSFM(using VLC to generate .jpg images from .mp4 files is an easy way to get a data set) and a closed source product with a free demo mode called "PhotoScan3D" from Agisoft. Others exist but these are the two I have experience with and am familiar with.

I don't know the extent of overlap between SLAM and SFM but I traditionally think of SLAM in regards to Laser Scanner data for mapping an indoor environment for the purpose of navigating it. Similar to what "Skanect" does although without the navigation.

Re:SLAM? (2)

fenris60 (925596) | about a month ago | (#47343955)

For me the interesting part is integrating SLAM with the object recognition and building semantics on top of that. Can you use the relatively powerful situation calculus to give the robot objectives? They seems to suggest this might be the case. For anyone with a Kinect and a fair bit of patience, you can try out an RGBDSlam algorithm for yourself (http://wiki.ros.org/rgbdslam).

Unfortunately (4, Funny)

mentil (1748130) | about a month ago | (#47343553)

They tend to bump into the same walls repeatedly before learning they're there and proceeding to bump into the adjacent wall.

Re:Unfortunately (0)

d'baba (1134261) | about a month ago | (#47343663)

So? Take it further or shut up.

Re:Unfortunately (0)

StankeyoSmith (3715761) | about a month ago | (#47343737)

So?

Take it further or shut up.

why so angry?

Re:Unfortunately (0)

Anonymous Coward | about a month ago | (#47346731)

Don't bitch at someone for posting a fact. Bitch at someone for posting a fact without proof.

Re:Unfortunately (0)

Anonymous Coward | about a month ago | (#47347721)

They tend to bump into the same walls repeatedly before learning they're there and proceeding to bump into the adjacent wall.

Just add a laser rangefinder and create an autonomous mapping drone like in the movie Prometheus. It might be useful for rescuers to have a map of the passageways inside of an earthquake collapsed building or coal mine.

What could possibly go wrong? (1)

cascadingstylesheet (140919) | about a month ago | (#47343905)

The more intelligent and autonomous my flying robots are, the better, I say ... hey, what are you ... Gahhhhhhhhh!!!

Prequel to Prometheus! (0)

Anonymous Coward | about a month ago | (#47344075)

This is how the "PUPS" the geologist in Prometheus used to map the cave (and the engineers ship).
So Ridley Scott just stole this idea from SLAM or the Sheffield university!!!!!!!!!!

Re:Prequel to Prometheus! (1)

mark_reh (2015546) | about a month ago | (#47344123)

As soon as I saw the article I thought that it was just the sort of thing they had in Prometheus. It would be extremely useful for the military to be able to map out the inside of a building. Of course, you won't know what's behind closed doors, but they'll add something for that next.

off the shelf software (3, Informative)

citizenr (871508) | about a month ago | (#47344133)

They are using PTAM package from Uni of Oxford
http://www.robots.ox.ac.uk/~gk... [ox.ac.uk]

Whats more they are using off the shelf ardrone-PTAM package

https://github.com/nymanjens/a... [github.com]

and replicating something done TWO YEARS AGO by Jens Nyman (from Belgian uni)

https://www.youtube.com/watch?... [youtube.com]

so W T F

Re:off the shelf software (1)

AchilleTalon (540925) | about a month ago | (#47344733)

I guess this shouldn't have made its way to /., this is an internal news publication and they don't claim having made any breaktrough technological advance by doing this neither. See this as an internal publication to publicize what is done at Sheffield University to undergraduate students.

Not only SLAM. (0)

Anonymous Coward | about a month ago | (#47345999)

Notice that this is not only locartion and mapping but also object recognition.

skynet? (1)

mpfife (655916) | about a month ago | (#47347847)

You want to get skynet? Cause this is how you get skynet.

Yay, another likely tool of war (1)

musth (901919) | about a month ago | (#47348445)

Not too easily purposed to warfare and domination of other peoples. Just what we need more of.

Re:Yay, another likely tool of war (1)

dywolf (2673597) | about a month ago | (#47349879)

Its just the evolution of terrain following radar.
No one here remembers the F111 family, capable of flying on autopilot through canyons at penetration speed, wingtips just feet from the walls?

Check for New Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...