Scientists Create AI-Powered Laser Turret That Kills Cockroaches 94
An anonymous reader quotes a report from Motherboard: Everyone wants to be able to just zap a bug and have it go away. But now, thanks to a recent development from Ildar Rakhmatulin, a research associate at Heriot-Watt University interested in machine learning and engineering, this dream is now a reality. In the study -- which was conducted last year but published in Oriental Insects last week -- Rakhmatulin and his co-authors used a laser insect control device automated with machine vision to perform a series of experiments on domiciliary cockroaches. They were able to not only detect cockroaches at high accuracy but also neutralize and deter individual insects at a distance up to 1.2 meters. This is a follow-up of sorts to earlier projects, in which he used a Raspberry Pi and lasers to zap mosquitoes. However, for this project, Rakhmatulin used a different kind of computer which allowed for more precision in detecting the bug.
"I started using a Jetson Nano that allowed me to use deep learning technologies with higher accuracy to detect an object," Rakhmatulin explained. The Jetson Nano is a small computer that can run machine learning algorithms. The computer processes a digital signal from two cameras to determine the cockroach's position. It transmits that information to a galvanometer (a machine that measures electric current), which changes the direction of the laser to shoot the target. According to the paper, Rakhmatulin tried this configuration at different power levels for the laser. At a lower power level, he found that he could influence the behavior of roaches by simply triggering their flight response with a laser; this way, they could potentially be trained to not shelter in a particular dark area. At a higher power level, the cockroaches were effectively "neutralized," in the paper's language -- in other words, killed. "I use very cheap hardware and cheap technology and it's open source," Rakhmatulin said. "All sources are uploaded in my GitHub and see how to do it and use it. If it can damage cockroaches, it can also damage other pests in agriculture."
It's not quite ready for household use though. "It's not recommended because it's a little dangerous," Rakhmatulin said. "Lasers can damage not only cockroaches but your eyes."
You can view a video of the device in action here.
"I started using a Jetson Nano that allowed me to use deep learning technologies with higher accuracy to detect an object," Rakhmatulin explained. The Jetson Nano is a small computer that can run machine learning algorithms. The computer processes a digital signal from two cameras to determine the cockroach's position. It transmits that information to a galvanometer (a machine that measures electric current), which changes the direction of the laser to shoot the target. According to the paper, Rakhmatulin tried this configuration at different power levels for the laser. At a lower power level, he found that he could influence the behavior of roaches by simply triggering their flight response with a laser; this way, they could potentially be trained to not shelter in a particular dark area. At a higher power level, the cockroaches were effectively "neutralized," in the paper's language -- in other words, killed. "I use very cheap hardware and cheap technology and it's open source," Rakhmatulin said. "All sources are uploaded in my GitHub and see how to do it and use it. If it can damage cockroaches, it can also damage other pests in agriculture."
It's not quite ready for household use though. "It's not recommended because it's a little dangerous," Rakhmatulin said. "Lasers can damage not only cockroaches but your eyes."
You can view a video of the device in action here.
go Hoihoi-san (Score:5, Funny)
One step closer to get Ichigeki Sacchu!! HoiHoi-san (One-Shot Bug Killer!! Interceptor Doll HoiHoi-san)
Re:go Hoihoi-san (Score:4, Funny)
Damn, just what I was about to comment. My kingdom for mod points. Or just consider this a "+1 Funny".
Re: (Score:2)
curious.
how easy it it to ignite dust bunnies
Take Care (Score:5, Funny)
Cover up that cockroach tattoo.
Re:Take Care (Score:5, Funny)
Cover up that cockroach tattoo.
Seems like a self-solving problem! Free laser tattoo removal for all cockroach tattoos!
Re: (Score:3, Informative)
Re: (Score:2)
He practically volunteered himself already.
Re:Border Deployment (Score:5, Funny)
Re: (Score:1, Funny)
The developer should have tested it outside the US Capitol on January 6th.
FIEND!
Do you have ANY IDEA how many FBI agents that would have killed?!?
Re: (Score:1)
Merrick Garland is a career criminal? Who knew?
Re:Border Deployment (Score:5, Insightful)
Right, so thousands of meters between devices with a range of meters, before you start looking at line-of-sight effects. It's knee-jerk reactions like this that makes a segment of the population SO MANIPULATABLE by any politician that promises to DO SOMETHING. The result would just be yet more of your tax syphoned off into the pockets of some politicians second-cousin.
Want to solve the 'problem' of immigration for less than you're currently paying? Either use the money to stabilise the country they are coming from, or use the money to set up schemes to integrate them, and disadvantaged locals, as productive members of society.
Re:Border Deployment (Score:4, Insightful)
That would mean actually solving the problem though which means they'd lose their wedge issue. They're better off with continuing to push for the wall, that wont solve a damn thing but it would make for some great photo ops if it were ever built. "Look at our bold new freedom wall!"
Re: (Score:2, Informative)
You do realize that defense in depth is a thing, and that the wall was part of the solution, not the whole thing?
I'd prefer the politician that tries to solve the problem rather than the politician that denies there is a problem.
Re: (Score:2)
You do realize that defense in depth is a thing, and that the wall was part of the solution, not the whole thing?
The problem is the depth. If it isn't deep enough, they'll go right under it, just like they did with the fence. And if it isn't comically tall, they'll throw a latter up against the wall, climb up, throw a ladder down on the other side, and climb down.
I'd prefer the politician that tries to solve the problem rather than the politician that denies there is a problem.
Nobody denies that there's a problem. They just differ in what they think the problem is. On the left, they think the problem is that legal immigration is too hard and that quotas are too low. On the right, they think the problem is too much immigration,
Re: (Score:3)
The problem is the depth. If it isn't deep enough, they'll go right under it, just like they did with the fence. And if it isn't comically tall, they'll throw a latter up against the wall, climb up, throw a ladder down on the other side, and climb down.
So, what you are saying is you have no clue what defense in depth means.
To enlighten you, defense in depth means that you have multiple defensive measures that when used together provide far more security than individual measures. In the case of the border, there was, the wall, but there were also other measures to protect against exactly the issues that concern you. There was to be video monitoring, radar detection, motion detection, and BPS patrols. Taken together, these measures would provide a much m
Re: (Score:3)
The problem is the depth. If it isn't deep enough, they'll go right under it, just like they did with the fence. And if it isn't comically tall, they'll throw a latter up against the wall, climb up, throw a ladder down on the other side, and climb down.
So, what you are saying is you have no clue what defense in depth means.
I've personally written documentation for a major computer company on that subject. Yes, I know what the term means. No, what I was actually saying is that I mock the very notion that such a tiny incremental change in an existing aspect of security qualifies as defense in depth. I used a play on words (talking about the depth of the wall in the ground) to do so.
Defense in depth isn't about replacing one security mechanism with an incrementally better alternative. It is about ADDING extra defensive measu
Re: (Score:2)
Defense in depth isn't about replacing one security mechanism with an incrementally better alternative. It is about ADDING extra defensive measures that complement the existing mechanisms. Adding drones to monitor the border is a great example of defense in depth. Replacing a fence with a wall is just defense, not defense in depth.
Exactly. The border wall construction was never about replacing a fence with a wall, there were many electronic security measures that were also being implemented.
The news media likes to mislead that it was only ever about building a wall, if you look at what was actually going on, and what was actually being built, it was a defense in depth approach including motion detection, cameras, patrols, and active radar installations. Arguing walls don't work, while ignoring the massive amount of security measure
Re: (Score:3)
Defense in depth isn't about replacing one security mechanism with an incrementally better alternative. It is about ADDING extra defensive measures that complement the existing mechanisms. Adding drones to monitor the border is a great example of defense in depth. Replacing a fence with a wall is just defense, not defense in depth.
Exactly. The border wall construction was never about replacing a fence with a wall, there were many electronic security measures that were also being implemented.
But again, why build the completely unnecessary and incredibly expensive wall when there was already a perfectly good fence? For that matter, why build permanently installed motion detector systems in a few spots when adding more drones would do the same thing for a fraction of the cost, and with far more flexibility for handling the inevitable situation where drug smugglers and human traffickers start crossing at a different point that is less fortified?
Oh, right. Trump couldn't advocate increasing the d
Re: (Score:2)
You do realize that thousand of migrants die each year https://www.statista.com/stati... [statista.com] crossing a literal sea to get from Africa to Europe. What the hell is a giant, massively expensive wall going to do for the US that the Mediterranean clearly cant do for Europe?
Furthermore, billions of dollars to build and billions more to maintain over time for a wall that people can climb over bare handed https://www.youtube.com/watch?... [youtube.com] isnt good value even if it was part of some sort of multi tiered immigration det
Re: (Score:2)
You do realize that thousands of migrants die in the desert crossing where there is no border in the US?
What does that have to do with anything I was talking about? As I stated, and you seem oblivious to, the wall was never meant to stand alone, so why do you focus on one component of border control, and ignore every other piece of it? The border wall has been proven effective where it is built,, the statistics show that the illegal entry through the wall is considerably less than the illegal entry where
Re: (Score:2)
the statistics show that the illegal entry through the wall is considerably less than the illegal entry where there is no wall
Hahaha. Of course most are going to go around through the path of least resistance, that doesnt at all mean we would get billions of dollars worth of illegal immigration deterrent out of it.
You really should learn what defense in depth means before spouting off arguing against something I didn't say.
Are you just going to tell that to anyone who dares disagree with you? We all know what you mean by defensive depth, it just doesnt matter relative to the point that the wall is a massive waste of
Imagine... (Score:5, Funny)
Re: (Score:2)
Room-boom.
Interesting, but bad idea (Score:4, Interesting)
This is the equivalent of a landmine. Sure, it's cheap and does what it's supposed to do, but it's indiscriminate. All it takes is some cockroach drop on someone's face and that will be the end of this project.
The ideal "solution" here is probably to just have the "roachtrap" only work with an infrared sensor to detect if anything as big as a cat or baby is in the room and disarm the laser. Once there are no potential accidents in the room, then activate.
Though let's be honest, I sincerely doubt this can be used safety. Imagine deploying this in a garage and it hits the gas tank of a car, or burns a hole in a car tire. There will probably be hundreds of pits burned into the surroundings as the roaches are killed with the laser. The ideal place to actually use this would be somewhere where no human will be (eg inside a wall, inside a crawlspace) and if we're going to go that far, have it kill dust mites as well to keep the crawlspace clean.
Re: (Score:3, Insightful)
Roaches tend to enter houses through spaces much too small for humans, or anything even close to a human. They also occupy spaces with the same characteristics. Plant a few of these things strategically in areas where humans would rarely encounter them and turn them off if exposed to direct light of any kind (roaches like dark places).
You would not put this device in the middle of your living room.
Re: (Score:3)
Strange realization: at the moment a cockroach is in my eye, I'd WANT to be shot in the eye with a laser...
Re: (Score:2)
Looking at the paper I think they are using class 3b and class 4 lasers. These are tightly controlled in even lab or f
I wish this had existed 20 years ago (Score:2)
I used to live in a student residence where this device would have been an absolute godsend. The problem is that it would have needed its own nuclear reactor to zap enough roaches to make a difference.
Re: (Score:2)
Wish it had existed 40 years ago, when I was going through the same thing.
On the other hand, a robotic death ray in the room would have been one of those things that kept me awake at night...;-p
Re: (Score:2)
Given some of the "digestive issues" triggered by excessive reliance on a diet of beer and pickled eggs, a robot death ray would have been trivial.
Re: (Score:2)
Re: (Score:1)
Re: I wish this had existed 20 years ago (Score:4, Interesting)
https://www.webstaurantstore.c... [webstaurantstore.com]
Ask and ye shall receive
Re: (Score:2)
Such a device could be useful over the long-term if only activated when humans were inside with the doors and windows closed. Let it zap the bugs over the course of months or years until the local population takes a nosedive. Bonus points if you could somehow lure a large number of the little bloodsucking bastards closer to the laser.
Love, Death & Robots (Score:3)
Re: (Score:1, Insightful)
Some engineer 100% watched that and said "I can do that!"
Or they just saw the laser that could hit mosquitos years ago.
Re: (Score:3)
Don't forget Baxter Stockman's Mouser robots from TMNT:
https://turtlepedia.fandom.com... [fandom.com]
Granted, they didn't have lasers . . .
Re: (Score:2)
I was thinking of the Andromeda Strain myself...
Paging Hollywood scriptwriters (Score:3)
Uh huh ... just wait. (Score:2)
Scientists Create AI-Powered Laser Turret That Kills Cockroaches
It's all fun and games until the cockroaches get hold of this tech.
Remember this one? (Score:4, Interesting)
Re: (Score:3, Interesting)
I was hoping that would become a thing, but it appears they were never able to get it to work reliably or safely. Probably because mosquitoes are very small and hard for a camera to spot and triangulate. You would need a high frame rate because they move so fast, as well as decent resolution.
It should be easier now because cameras and image processing hardware have come a long way. At the time I wondered if the task could be made easier by attracting mosquitoes to a "kill zone" somehow, a well lit area with
Re: Remember this one? (Score:2)
A limiting factor to reduce cost is that you really want to use eye safe IR pulse lasers, which while common are still relatively expensive. Also I doubt any corporate lawyer would sign off on US sales.
Put it in the sewers (Score:2)
That's where roaches generally live and breed, and its perfectly safe because nothing else is supposed to wander around sewer pipes. Make it kill mice and rats too.
Ever heard of exploding manhole covers? (Score:1)
You don't want this in sewers. Might ignite sewer gas. Boom! Flying manhole covers and gouts of flame in the street.
On the plus side, you might kill even more roaches and reduce greenhouse gas--methane burnt into carbon dioxide and water is less of a greenhouse gas than methane.
rat killing lasers in Project Wildfire (Score:2)
Where's the cat? (Score:2)
OH MY GOD...MUFFY!
Re: (Score:2)
Lower power level... (Score:3)
At a lower power level, he found that he could influence the behavior of roaches by simply triggering their flight response with a laser
Eww, no. I'm not squeamish around cockroaches, but even I hate when they fucking fly.
Slashdot, I'm disappointed (Score:5, Funny)
20 posts in, and no one's asked about mounting this on a shark?
Re: (Score:3)
Austin Powers, in 1999, was closer to the first IBM PC (1981), than we are now to Austin Powers.
Re: (Score:1)
But evolution? (Score:2)
Prepare for cockroaches with mirrors on their backs, Then, swarms of the blighters focusing the rays on their prey ,,,,
Shields up! (Score:2)
Watching the video, the laser seems to have started hitting the target cockroach at about 20 seconds in. It then took another 40 until the 'roach stopped moving. That doesn't exactly look like a humane instant kill.
Re: (Score:2)
Re: (Score:1)
it was only 1 time with the low-power laser. With a power laser not easy to make a video
As stated in the 1st line of the video description he used a low power laser to not saturate the video.
Oh good, more animal torture. (Score:1)
Disgusting.
Old idea (Score:3)
More than a decade ago, Intellectual Ventures developed a system that would kill mosquitoes with a laser. In mid-flight, and with the ability to sense the particular species it was going to target.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
Not long after IV came out with the zapper, I proposed using it to protect bees from the giant wasps. Could never get anyone from IV to talk about a license.
See that?! (Score:2)
You fucking cockroaches are NOT going to survive a nuclear holocaust!
That's right. We'll make sure of it. A Wind/Solar/radiation powered flying AI drone with an anti-cockroach turret.
I see (Score:2)
I mean still.
"This is a follow-up of sorts to earlier projects, in which he used a Raspberry Pi and lasers to zap mosquitoes. "
That project was abandoned after the scientist went blind for some reason.
Re: (Score:1)
it seemed like there was a small crackling sound a (Score:2)
Beyond the safety issues that everybody's already discussed can you imagine the stench?
Not ready for prime time (Score:2)
Turn it on at night . . . (Score:2)
Dog treats in the morning!
Needs more power ... one and done (Score:3)
The video showed the existing laser had to be kept on target for a while. Weren't you hoping for something more like 'pew-pew! dead pew-pew dead'?
The smell (Score:2)
AI controlled weapons... uh no (Score:1)
Today mosquitoes and roaches.
Tomorrow the high power version mounted on a military vehicle killing people. I'm sure nothing could possibly go wrong.
Re: (Score:1)
I'm not concerned about energy weapons. Getting melted leaves you just as dead as being shot with a standard round.
I am talking about the AI aspect (connected to any weapon it can fire at will).
A big ass truck with a bunch of soldiers with a laser can burn a hole in my chest. The same soldiers with a 50 cal can put huge holes in my chest. Either way I'm still dead. But I'm dead only if another human being consciously decides to aim their weapon at me and fire it. Hopefully I'm never in a position where
Why? (Score:2)
Why target cockroaches? They aren't a significant disease vector. Unless you leave your food lying around. And then that's true of most vermin. They serve a valuable function as scavengers. So their presence is an indicator of the poor hygenic practices of their human co-habitants. Which may be why they are disliked so much.
If you have cockroaches, you need to fo a better job of housecleaning.
Re: (Score:2)
You don't live in the South, do you! In the Gulf Coast and southern Atlantic states, cockroaches are in everyone's homes. They can't be eliminated, only managed. Worse, many of them can fly.
Aperture Science (Score:2)
Dare I say...? (Score:1)
Re: (Score:2)
What could possibly go wrong? (Score:2)
Just be sure your eyeball doesn't resemble a cockroach to this AI.
Evolution ... (Score:2)
Even if it works correctly (Score:2)
You're going to wake up to the smell of burned cockroaches every morning.
Re: (Score:3)
(Thanks for the opportunity to say that, btw)
Back in high school... (Score:2)
Channeling Walter Sobchak (Score:2)
Imagine.. (Score:2)
A beowulf cluster of these things!
new product soon on Amazon (Score:2)
Forget that video (Score:2)
Video too violent for Youtube? Tch tch tch. Damned cockroach lovers :-(