Live data from Hacker News

Predict 3D position of flying bugs and pest to shoot down them with a laser

naro.go.jp

81–90 of 124 posts

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#81

Intellectual Ventures made it seem promising enough in 2010, the slow-motion shootdowns are pretty amazing https://www.youtube.com/watch?v=BKm8FolQ7jw Guess there were some obstacles that couldn't be overcome

Supposedly the "Photonic Fence" underwent a big test in 2017:

https://nymag.com/intelligencer/2017/07/laser-shooting-mosqu...

"... The Photonic Fence will finally be tested later this summer in Florida, in a screened-in structure, against the Asian citrus psyllid, an invasive bug that is devastating the state’s orchards. So long as it manages to leave the bees alone — the last thing we need is more dead bees — it will then be tested in the open. ..."

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#82
post #77

Earlier quoted context omitted.

The laser is designed to shoot small things in the air. If you flip a coin, and the system has a bug that determines the coin is a bug then it'll shoot. Since you're surely planning on caching the coin, you'll look directly at it in flight. Now... you're blind.

That’s not how reflections work. Also there’s no bug as big as a coin that such a system could/should kill, so it’s pretty easy to not misdetect that. Again, corners have a higher probability of ending up in your eye than even just someone flipping a coin in 2022.

The comment said your system has a bug as in it malfunctions and decides to shoot the coin anyways, and his other point was you will be looking at the coin wide eyed so perfect chance for a reflected laser to hit you in the eye if by chance the correct angle happens to line up and send it back to your eye. Also, I've seen too much unbelievable stuff actually happen so nothing is unbelievable in 2022. Yes we are talking highly unprobeable but with 7 billion people on earth we have to account for the unprobeable to happen.

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#83
post #77

Earlier quoted context omitted.

The laser is designed to shoot small things in the air. If you flip a coin, and the system has a bug that determines the coin is a bug then it'll shoot. Since you're surely planning on caching the coin, you'll look directly at it in flight. Now... you're blind.

That’s not how reflections work. Also there’s no bug as big as a coin that such a system could/should kill, so it’s pretty easy to not misdetect that. Again, corners have a higher probability of ending up in your eye than even just someone flipping a coin in 2022.

It's just an example. I've worked with industrial laser cutters (100+watt CO2) before and even with all of the protective equipment built into the machines, you still wear glasses. The idea that you could use something like this in an exposed way is crazy to anyone who's actually worked with them. Any mistake is instant and permanent. You can't make a consumer product out of this. Maybe like a large box that only mosquitos could enter but at that point, you have easier methods than a visual system moving a laser to mm tolerances.

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#84

Earlier quoted context omitted.

Maybe if several low power lasers were fired together, to converge at the predicted position, it would moderate the risk of collateral damage. Even a few millimeters away from the target position, the radiative flux (W/m²) would be much lower, likely at safe levels.

What if all of those lasers converge to hit someone's eye? I don't see how a laser can have enough power density to kill a mosquito but still be eye safe.

We are talking about two different safety issues:

(1) Targeting the wrong thing.

(2) Even if the targeting is correct, missing by a small amount, or running the laser for slightly too long, and hitting something else by mistake.

If the lasers converge for full flux, the risks associated with (2) are better managed, but that doesn't speak to (1).

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#85
post #73
post #56

Earlier quoted context omitted.

After abandoning the laser idea due to the aforementioned “fully automated blindness / house fire machine” issues, I’ve often idly wondered if you could use a phased array of small subwoofers to generate a precisely targeted region of extreme pressure drop. That way you could just instantaneously and explosively decompress the suckers in such a way that’s probably safe for mammals.

There are a number of DIY instructions for building a phased array ultrasonic manipulator, eg [1]. You might be able to 'tweeze' the offender into a capture apparatus or high voltage deanimator. https://hackaday.com/2021/08/18/phased-array-levitation-is-s...

Why not the same idea with lasers, if you have an array of low power lasers but they all focus on one 3D location, if you are not at the focus location you will not be exposed to a dangerous laser power. An array of hundred to a thousand low end laser diodes using a an array of mems mirrors to aim the beams will ensure the focus point will fry the bug.

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#87
post #74

Earlier quoted context omitted.

I've been dreaming about that to kill mosquitoes since I was a teenager. Then I wondered: "how to prevent firing the laser into someone's eye?" and my idea was to only shoot laser from 2 meters high and to only ever shoot towards the ceiling, never downwards, while having the ceiling having some kind of paint/material that wouldn't be reflecting the laser too much. Make the paint/material all white to makes mosquito…

Wipe them out, all of them. The ecosystem will survive mosquitoes' extinction. On the subject of laser safety, can you make one weak enough that a human will naturally blink or look away without getting blinded but powerful enough that it'll overheat the mosquito its trained on for a few seconds? If the Joules are spread out linearly over a few second, I expect it'd exponentially decrease the risk of chance reflectio…

You shoot a nuke down a bug hole, you got a lot of dead bugs. — Ace Levy

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#88
post #82
post #77

Earlier quoted context omitted.

That’s not how reflections work. Also there’s no bug as big as a coin that such a system could/should kill, so it’s pretty easy to not misdetect that. Again, corners have a higher probability of ending up in your eye than even just someone flipping a coin in 2022.

The comment said your system has a bug as in it malfunctions and decides to shoot the coin anyways, and his other point was you will be looking at the coin wide eyed so perfect chance for a reflected laser to hit you in the eye if by chance the correct angle happens to line up and send it back to your eye. Also, I've seen too much unbelievable stuff actually happen so nothing is unbelievable in 2022. Yes we are talki…

And how many of those 7 billion people will die of mosquito-borne diseases...?

It's time to wipe out them out. We don't need mosquitoes, and neither does Mother Nature. Use gene tech to do the job, not Rube Goldberg hacks with lasers.

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#89
post #82

Earlier quoted context omitted.

The comment said your system has a bug as in it malfunctions and decides to shoot the coin anyways, and his other point was you will be looking at the coin wide eyed so perfect chance for a reflected laser to hit you in the eye if by chance the correct angle happens to line up and send it back to your eye. Also, I've seen too much unbelievable stuff actually happen so nothing is unbelievable in 2022. Yes we are talki…

And how many of those 7 billion people will die of mosquito-borne diseases...? It's time to wipe out them out. We don't need mosquitoes, and neither does Mother Nature. Use gene tech to do the job, not Rube Goldberg hacks with lasers.

Oh I am all for laser defense, I was merely pointing out that op is correct in that a system like this will likely go wrong some how. But I honestly don't think they would use them in household setting but rather use them on the outskirts or around ponds and use it for other pests in fields away from human eyes.

Re: Predict 3D position of flying bugs and pest to shoot down them with a laser

#90
I wanted to do something similar but it was related to an idea I wanted to also try. The project was a dashcam that could determine the velocity of other vehicles. Some implementation of OpenCV is what I thought may work. It was the triangulation part I wasn't sure about which I think is vital to get it to work.
Post reply on HN