h a l f b a k e r yStrap *this* to the back of your cat.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
A drowning person will almost invariably struggle for as long as they can while making almost no forward progress before being overcome by exhaustion and slipping under.
Idea: Position water level detection panel "tiles" strategically around the pool at the water level and analyze the waves with a
computer.
Number crunching on enough data points would reveal intensity and location of the source(s) of the waves.
If wave patterns suggest drowning, then alarms and other life guarding mechanizms would be activated.
It would take some good programming and lots of sensor tiles to be able to distinguish between different situations involving different numbers of swimmers and their pool behaviors and (of course) it would have to take the 3-d shape of the pool into account. The system could even be set up to calibrate itself. (Pool party!)
This would do nothing for people who drown after spinal cord injuries, but for your every-day run-of-the-mill "swimmer gets a cramp" or "kid falls into the pool" scenarios it could save lives.... It could even be set to detect problems for swimmers prone to seizures who otherwise might be unable to swim safely.
Big Brother of the Pool
http://www.poseidon-tech.com/us/unit.html
Wonder if it runs MS Windows... [Don Quixote, Dec 13 2004]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Destination URL.
E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)
|
| |
Perhaps Cray would find a new market in this. |
|
| |
Wouldn't it make more sense to use cameras to watch the pool? Such a system is in development, I believe. I imagine it uses similar technology to station-platform units that watch for suicides. Apparently their behaviour pattern is fairly distinctive, such that they can be identified and stopped before they throw themselves off the platform. |
|
| |
<edit> Such a system is in production, Don Quixote believes. He's right, too (see his link). |
|
| |
Somewhat baked -- though the system watches the swimmers (rather than the waves) and doesn't use high-tech tiles. Linky... |
|
| |
Most baby / toddler pool drownings happen silently and with little surface 'struggle'. |
|
| |
These gizmos may give a false sense of security - the central message of pool safety is that no child should be inside the pool fence, unobserved by an adult, for even a second. |
|
| |
Alas, (I seem to like this word :D) idioscism strikes again (no offense). First: they already have a system for this. It's called cameras. The military base just outside of town's pool has two of em'. |
|
| |
Second: Any kid could just continously splash water on the tiles and trigger the alarm for fun. |
|
| |
Third: The kind of programming for this is virtually impossible right now. It would require a form of AI (artificial intelligence) which we have no way to make at this point in time. The system would be setting off alarms randomly throughout the day! Cameras would be cheaper than this. |
|
| |
Oy... it is impossible to make such a system. Todays AI (they use this type of AI in video games, which is scenario based) is scenario based. Like this: |
|
| |
You are an AI. You are being shot at by a player. You are backing up against a 3D wall, while shooting back at the player. Choices: |
|
| |
B. Charge straight at the player and attempt to use your melee attack. |
|
| |
C. Throw a plasma grenade and attempt to make it stick on him and detonate, killing him instantly. |
|
| |
If you tried to do this in real life, your program would be infinitely long. Bedcause ANYTHING can happen in real life, it is impossible to do this. |
|
| |
Yes, yes. You are all right... it IS impossible.
Could never be done.
We should just stick with what's already invented. |
|
| |
It's impossible, and couldn't exist. And to boot: it's already baked...... |
|
| |
Just paying someone to watch your pool (on camera and in real life) 24-7 is a MUCH better idea. |
|
| |
Come on people. The bakery isn't about what could actually be built today, it's about ideas. Many, many invented things were thought to be impossible right up to the day they were demostrated. |
|
| |
Pattern recognition is used in other crisis sensor applications. For example, there are new residential circuit breakers which detect the distinctive electrical pattern of an arcing fault, which often does not have enough current to trip a standard circuit breaker. There are arc FLASH sensors in industrial equipment which look for the visible light spectrum associated with arcing faults. (Guess what I do for a living??) Anyway, pattern recognition is a good approach to many problems. Croissant (don't drop it in the pool!) |
|
| |
[EvilPickels] you seem to type without writing. |
|
| |
People drown differently and swim & play differently. It would be possible if there were any sort of "pattern" to how drownings occur versus non-drowning pool use. We may someday invent better pattern recognition capabilities, but I don't think we'll ever have patterns to active drownings, so [-]. |
|
| |
But, you could detect people who've already drowned perhaps..... |
|
| |
There are sonar patterns to detect the size, shape, & orientation of fish in water, but that's a shape pattern, not a motion pattern. |
|
| |
Rather than detecting the level & splash, you might have sonar detection of the body who's fully underwater for more than 2.5 minutes, and then send in the rescue before brain damage is permanent. And that also means you need a "lifeguardless rescue method". |
|
| |