Buildings Could Save Energy By Spying On Workers

Galactic_grub writes "In the future, your place of work (or apartment) may very well spy on you. But that doesn't mean it'll be able to name and shame you for all your nasty habits. Researchers at Mitsubishi Electric Research Laboratory (MERL) have devised a 'dumb' surveillance system that monitors the movements of workers without identifying them individually. The idea is to have a computer system automatically configure the air-conditioning to save money, or illuminate the most appropriate escape signs in an emergency."

Comments

Emergencies?

[msauve]

illuminate the most appropriate escape signs in an emergency.

Well, I would illuminate all escape signs in an emergency, infinitesimal energy savings be damned!

Re:Emergencies?

[caffeinemessiah]

Which is why you don't work at MERL. In large, complex buildings, having only the signs that lead you out of the building in the quickest possible way would definitely be a benefit. Especially if, for example, one particular exit route was blocked for any reason.

Re:Emergencies?

[PhrostyMcByte]

I hope by "appropriate" they mean if a building is on fire and one of the escape routes like a stairwell is on fire, it will redirect them to a safer place.

Re:Emergencies?

[vux984]

Better yet if they can put enough AI into the system it might learn to create emergencies just to evacuate the building allowing it to shut all the lights and air conditioning off, and saving the company some real money!

Re:Emergencies?

[owlstead]

I've run straight towards a mayor gas leak in an actual emergency. If I had to do it all over, a smart sign that would get me away from the gas leak would be very welcome. Of course, the gathering point (for counting the employees) was also next to the gas leak, but trust me, I wasn't there for a very long time.

Re:Emergencies?

[Kjella]

Fair enough, if they have the sense to make it so that the default is ON. Lost connection somehow, sensors not responding for any reason? It lights and stay lit. And if they really went to the trouble of making it an automated system, they should have colored bulbs. Green = Primary exit. Yellow = Secondary exit. Red = Blocked exit. No light = BULB IS BROKEN. I mean seriously, if people don't use an exit because they assume they're not suppose to use it because it fails to light up, you've made a much bigger screw-up. In most cases there's some obvious signs this exit isn't usable, like flames and smoke...

Re:Emergencies?

[ShieldW0lf]

From the article:

Such systems could use their knowledge of where groups congregate to turn down the air conditioning when there are only a few people in one part of the building, for example. In an emergency, electronic signs could direct people to the nearest available escape route when one becomes congested.

Low-tech signs to indicate all exit methods. High tech signs to indicate the best exit method, taking into account that there are 200 people trying to jam their way out the closest exit, and one of the two alternative exits is on fire.

Re:Emergencies?

[seaturnip]

Fair enough, if they have the sense to make it so that the default is ON. Lost connection somehow, sensors not responding for any reason? It lights and stay lit.

Still too failure-prone. What if the sensors are responding but buggy and they actively tell the signs to be off? A major disaster requiring evacuation could cause all sorts of things to go wrong.

The most robust systems are usually the most simple.

Re:Emergencies?

[Anonymous Coward]

As I recall, some of the survivors in the WTC didn't take the "best" route at all.

These are emergencies we're dealing with, folks. All bets are off.

This is one Federal regulation I agree with. . .let me at least know where I'm going if I'm fending for myself.

Re:Emergencies?

[eonlabs]

It's alright if it fails to on. The issue here is when you have large office buildings (e.g. 50+floors) and nobody's there. The savings come from not having those lit 24-7 for weeks on end when there is no emergency.

It's the age old question:
If a tree falls in a forest, and no one's there to hear it, does it make a sound.

In the same breath, I would probably wait on that until the other modes of operation are proven effective (and not something remeniscient of 1984).

Re:Emergencies?

[Mr. Freeman]

He said "All ESCAPE signs" Not "all signs". Furthermore, only illuminating the nearest escape signs presents a problem if that one route is blocked.

Re:Emergencies?

[guaigean]

And if they really went to the trouble of making it an automated system, they should have colored bulbs.

And what do you propose for the very large population of color blind people?

Re:Emergencies?

[HermMunster]

Well, the concept is bullshit. If they can spy on you they will. Period.

Re:Emergencies?

[Who235]

And what do you propose for the very large population of color blind people?

http://www.tshirthell.com/store/product.php?produc tid=559

I know it's OT, but man, is this shirt funny.

Re:Emergencies?

[Hijacked Public]

Many modern manufacturing control systems use integrated safety systems. Some gigantic machine is being run with a PLC that tells motors to start when buttons are pressed, etc. The same PLC is also watching to see if a safety cord is tripped by some guy who is caught in the machine. This is pretty reliable if properly implemented.

For Exteme Super Safety you add a second PLC looking at the a second set of contacts in the safety cord. If at any time both PLCs don't agree, the machine stops.

This kind of setup is reliable enough for giant stamping presses, I suspect it would be OK for lighting up signage.

Re:Emergencies?

[Anonymous Coward]

How about if the system detected a fire in the stairway, it would turn off the Exit sign above that door? That might save a bit of time.

Of course, this would be using MORE electricity, but it would still be useful IMHO.

Re:Emergencies?

[dgatwood]

The biggest problem, IMHO, with emergency evacuation is a little fact that is often overlooked: people tend towards exits with which they are familiar. If you have two main exits and four emergency exits which trigger an alarm, nearly all people will automatically take the two main exits because they are comfortable entering and leaving the building through those entrances and exits. The four emergency exits will almost never get used. The only exception is a situation where you have to pass right by the emergency exit to get to the main exit, and even then, most people will be reluctant to use the emergency exit.

Similarly, stairs that culminate in an emergency exit door will almost never get used because people usually won't use them to get between floors and can't use them to enter or leave the building. As a result, only a small percentage of people will know where they are, and even fewer will know those stairs well enough to take them in an emergency. The rest will be too afraid of getting turned around and going the wrong way.

Making this even funnier is the corollary: nearly every alarm resulting from use of emergency exits is a false alarms. (A large percentage are either A. shoplifters at department stores or B. small children.) That makes the automatic alarm a complete waste of resources. You are far better off with a normal door and a pull handle.

What does this mean for emergency preparedness? Simple: you should NEVER have an entrance or exit that is "for emergency use only" unless it leads to an unsafe area such as a rooftop. All exits should be clearly marked and USED REGULARLY. Workers should be encouraged to frequent the entry/exit stairs closest to their offices/cubicles whenever possible, and to enter and leave the building through the most direct route, which by definition means that the most direct route cannot have one of those stupid "alarm will sound" bars on it.

If every emergency entrance and exit were turned into a normal entrance/exit with a pull handle and were used in the normal course of entering and leaving buildings, the typical time to evacuate a building in the event of an emergency would drop dramatically. If you'd like to take the easiest possible step towards making your workplace safer and you have an emergency management team at your workplace, point this out to them, and ask them why safe entrances and exits are marked as being for emergency use only. If they can't answer that question, ask that they change the doors into normal entrances and exits, then continue to nag them at every opportunity until they do so. You may not successfully change anything, but at least you'll get them thinking about the problem.

Re:Emergencies?

[PPH]

And if they're all blocked, the signs say, "You are going to die".

Re:Emergencies?

[Redlazer]

Truly awesome ideas.

Mod parent up.

-Red

Re:Emergencies?

[FFFish]

I agree. All emergency exit signs should be On at all times.

Blinking the most-appropriate exit path would be appropriate, so long as in the worst-case they failsafe to steady On.

Re:Emergencies?

[sumdumass]

Being reliable when properly implemented on equipment the operator receives instructions on as well as warning and such that they know before hand that it is life and limb threatening to operate is a different story entirely when considering someone stopping by a 50 story office building to get their eyes checked on the 29th floor.

I mean seriously, Unless your in a position to instruct everyone about the workings as well as the dangers and account for most every situation, it isn't really a good Idea. The dangerous machines have the safeguards in case of accidents and can be operated without likelihood of injury by a fully competent operator. It is a controlled environment were and escape route in case of fire or some other danger is in no way close to the same controlled situation. I'm not doubting the abilities of the sensors, I am doubting the ability for them to work properly with random human interaction in such an uncontrolled environment.

Re:Emergencies?

[Anonymous Coward]

How dare you attack colored blind people!

And if building walls were transparent we wouldn't need to spend money on lighting. So how about glass walls for the women's restrooms? And how about more energy-efficient secret webcams in the women's restrooms in federal buildings?

-- Signed, Bush Secretary of Energy

Re:Emergencies?

[sumdumass]

Two things. I agree with everything you said except that it isn't always financially feasible to place necessary staff at certain exits and entrances like in a store or a building that requires security. But these are far and few in between and outside of this, I don't see any problem in most situations with that idea.

Well, one more issue, Stairs and corridors that are rarely used tend to be haunted by seedy people who are prone to illegal acts. Rape, muggings and random violence seems to happen in these places. It could be considered part of the building security. I know of one apartment complex that has a back entrance leading from an alley that has become a hangout spot for drug users and what appears to be homeless people. It was the sight of several robberies and a few rapes, the cameras were defaced and rendered useless the day of the acts. They closed the entrance to all but emergency activity and the problems almost ceased to exist. There still is a lot of problems in the alley but not at this building.

And yes, the cops regularly patrol the area with this going on sometimes directly after they go threw it (withing 10 minute of when the patrol car went by). It is just a bad neighborhood with a few crack houses I guess and a mixture of people who don't care with some who care a lot.

Re:Emergencies?

[Firethorn]

Even for the practically blind guy, wouldn't it be better to avoid confusing him by not lighting exit signs for a route that leads through the danger? Most places require redundent exits from areas larger than 'Joe Manager's corner office'.

If one is blocked by fire, shut the lights off so people don't attempt to go that way. Use standard redundant wiring for it, to include test sensors to tell you when the thing goes bad.

Re:Emergencies?

[asolidvoid]

Though it doesn't take it to the degree you are suggesting (if I recall correctly) the Universal Building Code, applicable in most states, stipulates that for high occupancy spaces the main entrance path has to be sufficiently sized to handle at least half of the egressing occupants. Maybe this isn't enough, but I don't know the statistics on how people actually behave during an emergency.

Re:Emergencies?

[nwbvt]

Except if the route with the lit signs is also the route that is blocked.

Also, most large office buildings house a lot of people, if everyone just went out what some algorithm considered the quickest door there would be a huge bottleneck in that exit. Generally the plan is to make use of each and every available exit.

Re:Emergencies?

[sumdumass]

As long as the fire doesn't cut power or control and then end up directing people into danger after the system was designed to take them away. As it is now, the user needs to posses certain skills to ascertain the dangers in their exit route. These skill are taught in elementary schools and during the hiring process. There are also emergency evacuation layouts with the most direct exit outlined that needs to be in most buildings.

When we don't need these and stop paying attention because some other system is supposed to be better, it better be good enough not to kill someone. You would need a way to send signals even when the cables are burnt and the electricity is off. You need a failsafe in almost every aspect. Otherwise, it is just killing someone because it could mislead them into danger.

Re:Emergencies?

[Nutty_Irishman]

That depends. If you're a good employ that carpools, doesn't turn on the lights, and always powers down his machine at the end of the day, then you're fine. If you don't, well, just remember, stop drop and roll, stop drop and roll.

Re:Emergencies?

[dgatwood]

As far as the staffing goes, those issues can be easily handled by badge access doors, just like most businesses do for after hours access. With respect to the seedy characters outside the doors waiting to mug you... yikes! Fnd a job in a safer neighborhood. :-)

Re:Emergencies?

[Anonymous Coward]

What does this mean for emergency preparedness? Simple: you should NEVER have an entrance or exit that is "for emergency use only" unless it leads to an unsafe area such as a rooftop. All exits should be clearly marked and USED REGULARLY.

Fully agree, but for different reasons. We had an office where the emergency exit lead straight into the parking lot, where as the main entrance required people to walk around the building. Of course most people used the emergency exit, with the result that we got warned that the emergency exit is for emergencies only, and not to be used in other situations.

What happened was that we found out that some of the screws holding the stairs in place were missing. The stairs were unstable when used by more than one or two people at a time. Because we actually used the emergency exit, this was noticed and fixed quickly, where as if we had followed the rules and not used it, noone would have noticed, and the stairs would still be ready to fall down in an emergency when people don't have time to wait until noone is on the stairs before walking down.

Re:Emergencies?

[dgatwood]

I do. I've talked to some of the emergency management folks where I work. I believe that one or two percent is a highball estimate, and a real-world guess is more like half a percent. If the main steps were only big enough to handle half of the building, roughly half of the occupants of the building would burn to death, and frankly, that's being optimistic.

Those building codes definitely should be rethought. Desperately. Fortunately, at least in our case, we are still able to evacuate relatively quickly (not as quickly as they would like, of course) because the main steps have relatively high capacity. If the main steps caught fire, though, things could get a little rough. I think that people would get really, really confused.

Re:Emergencies?

[houghi]

Make it easy. Just send out an email with the procedure to allstaff@company and everybody will read it and be informed.

Re:Emergencies?

[Anonymous Coward]

And if building walls were transparent we wouldn't need to spend money on lighting. So how about glass walls for the women's restrooms? And how about more energy-efficient secret webcams in the women's restrooms in federal buildings?

It would probably be more efficent to stop sex segregation of toilet facilities.

Re:Emergencies?

[ajs318]

The problem with this is that doors which are designated as "emergency exits" and fitted with "dead man's bar" latches are usually not industrially-rated. The door companies know that they aren't going to be used that often, and so use cheaper materials and components (even residential doors are heavier-duty than emergency doors). Management get to save a few bob and impose another stupid rule (emergency exits are not to be used except in an emergency). Everyone's a winner!

I know this because in the factory where I used to work, one of the buildings had an emergency exit which, for many workers, was much more convenient than the main doors. This door actually broke from over-use and the firm refused to replace it, saying we had invalidated the guarantee by opening it too often!

Re:Emergencies?

[ajs318]

I can just see Clippit popping up on someone's screen .....

It looks like you are on fire. Would you like to

( ) Try and put it out
(o) Search MSN for more information on firefighting
( ) No thanks, I'll deal with it later

Re:Emergencies?

[guruevi]

I don't know about where you live, but in most factories I have been to, the way I learned it in school (yes, I have studied PLC controllers) and I think even governments require emergency stops, cords and the like to be switches serial to (one or all of) the main power lines. Somebody trips any emergency stop or the emergency switch breaks and the power gets cut off to the machine (not necessarily the controller, just the power that juices the motors for example).

Nobody in the right mind would let a programmer or a program make the decision whether or not an emergency switch has been activated, they are too prone to bugs and have too many points of failure either from the programmer that made a mistake or a bug in the firmware or even hardware that breaks.

Re:Emergencies?

[kabocox]

Which is why you don't work at MERL. In large, complex buildings, having only the signs that lead you out of the building in the quickest possible way would definitely be a benefit. Especially if, for example, one particular exit route was blocked for any reason.

Um, the building is on fire, or in an earth quake, flood, hurricane or what ever, it isn't working properly and is about to fall down. Are you going to trust that these signs will works as said? I doubt they'd work quite that well.

What we really need is inside of a federal holiday, a federal disaster drill day or week. We seperate the US into regions and drill the entire community into what to do if the most likely natural disasters happen there. Usually, hospitals, fire departments, police departments and the Office of Emergency Management have these sorts of drills every now and then, but only first responders are really drilled into what to do.

Do you remember in public school having fire and torando drills? Something like that on a city level. I don't have a clue how you'd drill a city like NY for everyone to just know the best way out of any given building, but that's really what we should do.

Re:Emergencies?

[sumdumass]

Fnd a job in a safer neighborhood. :-)

Well, to be honest, that situation wasn't there when it all began. It happened over the course of about 3-5 years. After the second rape and probably the 15th mugging with the cops unable to do something about it, the land owners in the area banded together and started buying the residential houses behind the area and bulldozing them to create a large parking lot.

I'm sure this not only displaced innocent people but moved the problems somewhere else so they didn't have to worry about them. Locking the doors made a big difference in the few years before this could happen though.

Re:Emergencies?

[Joebert]

Which is why you don't work at MERL. In large, complex buildings, having only the signs that lead you out of the building in the quickest possible way would definitely be a benefit.

Well, I don't work at MERL either, but uh, wouldn't be more efficient just to get rid of the wild-goose-chase exit signs all together ?

Maybe I'm missing somthing here, but I thought for sure that the whole purpose of exit signs was to illuminate the quickest way out of whatever room you're in to begin with.

Re:Emergencies?

[markov_chain]

It would be interesting to see a study of how many people actually use those EXIT signs in emergencies.

Re:Emergencies?

[Joebert]

Blinking the most-appropriate exit path would be appropriate, so long as in the worst-case they failsafe to steady On.

Unless it's Friday, then there's a good chance someone's going to mistake the smoke, blinking lights, & people screaming, for Casual Friday.

Re:Emergencies?

[Firethorn]

A good wiring job will prevent the lines from being cut for quite a while, and like others have said - if you loose the link, have them all fail over to 'on'.

The right wire in conduit can take quite a bit of heat before it fails as well.

Fire alarms for big buildings are already linked, this is just an extension of that.

Of course, my bar is quite a bit lower. I don't expect perfection, I simply expect reasonably better performance. Back this up with drills and other tests.

You might end up with fire alarms signs with directional speakers that go 'This way to the exit', or 'Warning: Fire in this direction', along with 'Good' exits being lit up even more.

Re:Emergencies?

[Hijacked Public]

I have been in dozens and dozens of factories all over the world, I used to work for a fairly large controls house.

10 years or so ago pretty much all of the electrical safety devices were as you describe. Most occupational safety organizations have changed with the times and now approve various integrated systems that are normally controlled with the same tpye of PLC (often the very same one) that runs all of the other field devices.

I've only worked with Allen Bradley stuff, so this would generaly describe these kinds of safety systems as I know them.

They are quite reliable. As for the programmer making a mistake, this stuff generally gets well tested before the equipment is released for production use. With the dual PLC setup I described above I have even seen companies hire two different controls houses to write the logic so that the same tasks are performed differently.

Re:Emergencies?

[Hijacked Public]

Safety signage in buildings is usually a matter of code complaince. In the US there are requirements for Exit signs, for instance, that specify the size and shape and typeface and the minimum lumens....etc.

This is so no one needs specific training on how to identify an exit when they are in some unfamiliar 50 story office building getting their eyes checked.

What I thought we were discussing was the automation system used to do various things throughout the building in the event of an emergency. Transient occupants of the building don't need to be trained on it. If the automation is good enough most of the employees, even the security people wouldn't need to know about it. Indeed, most newer large office buildings in the US (and a whole lot of them in Japan) are already heavily automated.

Essentially, if the system sees an output from the fire alarm, light up all the emergency lights. Or something like that. More likely, if the system doesn't see that the fire alarm is sending it an "all OK" signal, light the lights. No employee involvement or training necessary and certainly none for the guy getting his eyes checked.

Re:Emergencies?

[ffflala]

If every emergency entrance and exit were turned into a normal entrance/exit with a pull handle and were used in the normal course of entering and leaving buildings, the typical time to evacuate a building in the event of an emergency would drop dramatically.

That's true, but there are a number of reasons to limit regularly-used entrances and exits, for example secured buildings that need to authorize or screen all who enter. The more entrances/exits, the more difficult it becomes to secure a building.

Re:Emergencies?

[Anonymous Coward]

I preffer to switch the lights off by myself for now... don't know about the emergency yet... Although something like that happened when i was working as an operator for a small VOiP net in home country. There was a power outage, the alarm whent on i panicked the private cops came and everything was like a movie man! :)

Re:Emergencies?

[baggins2001]

I hope by "appropriate" they mean that if one of the escape routes is blocked by a gunman it will redirect me to a safe place.

Re:Emergencies?

[sumdumass]

You see, a fail over to always on is were the problem is for me.

If I go into a building and you tell me that in case of fire or whatever emergency, the most direct exits will light up leading you away from danger therefore to safety, and then all of them are on leading me into danger because of some glitch or wiring was burnt, whoever told me that I would be safe by following the lighted exist has now caused any injuries or deaths that might happen. Of course I still need to look listen and feel to see if there is danger on the other side of the door and stuff. But if i'm still doing this then whats the point of the other system?

It isn't like complete strangers end up in a building fighting for their life without anyone who is familiar with the building. Each office usually makes sure it is cleared before the employees themselves leave. If your willing to take the responsibility of all injuries and deaths because of failures in the system, By all means implement it in your building. I look at all emergency exit signs when entering a new building because they just catch our eye like they are designed to do. They are battery operated because on of the first things to go is the power (Well not first but common and another reason you don't use elevators during fire alarms.)

Re:Emergencies?

[Firethorn]

If I go into a building and you tell me that in case of fire or whatever emergency, the most direct exits will light up leading you away from danger therefore to safety, and then all of them are on leading me into danger because of some glitch or wiring was burnt, whoever told me that I would be safe by following the lighted exist has now caused any injuries or deaths that might happen.

What are the odds of the system failing before people successfully evacute when the system is properly designed and installed? Basically, in case of a fire, does the system, on average, save lives or cost lives? If it works the vast majority of the time and direct people to a good exit, it's better. If not, oh well.

What I'm envisioning is that the 'fail on' would be what's the standard today - battery backed up constant light. I'd have the 'directed path' be done through other signals, either an additional sign/light, voice, whatever.

I sometimes work in a building that's effectivly a maze. I've ended up following the exit signs simply to find my way out.

Re:Emergencies?

[sumdumass]

What are the odds of the system failing before people successfully evacute when the system is properly designed and installed? Basically, in case of a fire, does the system, on average, save lives or cost lives? If it works the vast majority of the time and direct people to a good exit, it's better. If not, oh well.

The odds don't matter, What you have done is taken the responsibility of directing the person to safety over a normally functioning system that lets the person at risk make those decisions. Think of it as someone wanting to commit suicide, in one hand they have a gun waiting to blow their brains out, If they give you the gun and ask you to do it, it is murder plain and simple even though they wanted you to do it.

What I'm envisioning is that the 'fail on' would be what's the standard today - battery backed up constant light. I'd have the 'directed path' be done through other signals, either an additional sign/light, voice, whatever.

This wouldn't bee too bad as long as the old system is still in place and functioning and people are instructed on what to do with it. And the Fail on would need to present something so as the user lost in a smoke filled room can know with little effort that it isn't guiding them to safety.

I sometimes work in a building that's effectivly a maze. I've ended up following the exit signs simply to find my way out.

So you know first hand how effective the current systems are. While not completely fool proof, it does allow you to make decisions concerning your safety instead of dictating these decisions via some system that could fail and lead a person into danger. The big difference is did my Aunt Martha make the wrong turn or did your device tell her to turn that way which eventually caused her to die. If it is your device, you just killed a person. If it was her old age, then she got herself killed. And I'm willing to bet that a bunch of lawyers would be happy to figure it out for us and I would own a building.

Re:Emergencies?

[Firethorn]

And I'm willing to bet that a bunch of lawyers would be happy to figure it out for us and I would own a building.

It's more complicated than you think. If such a system gained enough testing and studies to show that it's, on average, safer than the old static systems, then the bunch of lawyers could sue somebody for having only the old system. After all, I didn't take 'reasonable' measures to ensure the safety of the building's occupants in case of fire.

So you know first hand how effective the current systems are.

At least for that building, the current 'system' mostly consists of glow in the dark signs saying 'Exit' with an arrow. One good thing is that they're mounted about two feet above the floor, increasing the chances you'll see them through smoke. I'll note that this building also has effectivly no windows.

Re:Emergencies?

[sumdumass]

It's more complicated than you think. If such a system gained enough testing and studies to show that it's, on average, safer than the old static systems, then the bunch of lawyers could sue somebody for having only the old system. After all, I didn't take 'reasonable' measures to ensure the safety of the building's occupants in case of fire.

Currently the law describe what the reasonable measures are. As long as you are in compliance with building codes and fire-codes for the area you have an exemption of liability for anything that isn't directly your fault. Until this changes, only stuff like padlocked emergency exits or non working alarms and the likes are were you can get hit unless you actually started the fire. With a system that tells people were to go, you have now become responsible for their safety and I can think of no reasons someone would want to do that outside giving them the tools necessary to save themselves as the current law proscribe.

I don't think this can be treated like a unsafe product. There are laws specifically spelling out safety requirments and standards and as long as you meets those, your safe from these types of suits. The anti gun groups recently sued gun manufactuers over placing restricting devices on the guns that allow only the owner to use them. The gun manufacturers won because they already meet the safety reuirments and local ordinances state they have to be locked up and such.

And I seriously doubt it can be much safer then the existing rules. The existing system is pretty effective and the same problems that plague it will haunt this new thing too. When people are cut off from their exits because of smoke or flames, this cannot help. When people don't take the warnings seriously and leave before the floors collapse, this cannot help, when people panic and trample someone who has fell, this cannot help (which reminds me of an interesting story about a moshpit at an Megadeath concert). This is were the existing system breaks down and would be the same with any system. About the only thing this could help with is telling emergency response members were people are still trapped or were they were last seen alive.

At least for that building, the current 'system' mostly consists of glow in the dark signs saying 'Exit' with an arrow. One good thing is that they're mounted about two feet above the floor, increasing the chances you'll see them through smoke. I'll note that this building also has effectivly no windows.

Yea, I have seen quite a few different spins on stuff like this. I think it is more or less different local ordinances that bring about the small changes. I don't know if glow in the dark stuff would be visible in a smoke filled room but your not always going to be in smoke when leaving a building on fire. I had two clients in a building that was an entire city block long so you can imagine the lack of windows on the interior offices as well as the maze of corridors like you describes. These buildings had similar devices like you described too.

I guess my biggest problem with telling someone were to go durring an emergency is simply, if you direct them to danger, you could have killed them. This shouldn't be taken lightly in the least. Being legally and/or morally responsible for someone's injuries or death isn't a good position to be in Eventually, anyone controlling the device will have to leave, and human or machine, they are prone to making mistakes. This is why I think the current system is best were the people at risk make their own mistakes and are effectively holding their injuries and deaths in their own hands. Would you be comfortable knowing that something you administered or something installed in your building caused the deaths or injuries of several people because of some unknown bug caused by the fire, earthquake or whatever happened to trigure the emergency?

Re:test comment

[Anonymous Coward]

Ok.... Doh!

Already exists

[fwice]

My company already has this. Lights / ventilation run only from 6:30 AM to 6:30 PM. At that points, the lights / hvac goes out. But if you are working and want to reactivate these systems, there is a webpage (and a phone extension) that you go to, input your office location, and voila -- the systems come back on.

and according to the company, its dropped costs by a third.

Re:Already exists

[shayborg]

We have something similar, but a little less cumbersome. All the lights and office climate controls go out at 9 PM. For each block of a few (four or five) offices there is an override switch that you can hit to turn everything in that block back on for the next two hours.

Re:Already exists

[Smidge204]

I've been specifying "occupancy sensors" in commercial HVAC systems for years. Not only do they control the lights (so they turn off when someone leaves the room without hitting the switch) but they tie into the building management system/central HVAC control to handle things like ventilation cycles and temperature setbacks. Combine with a time clock like you describe and you get a great way to improve building efficiency. (No crappy website or phone call to deal with, though!)

The key difference in the article is, instead of just saying "Space is occupied" or going by time, it actually tracks individual persons and can tell how many people might be in a room or, more impressively, where a person might be going so it can plan ahead.
=Smidge=

Re:Already exists

[TheRaven64]

Lights in my building don't have an 'on' state, they are either off, or PIR-controlled. When they are PIR-controlled, they go off after no movement for about 10 minutes. You can see who is working at night by which bits of the lab have little pools of light.

Re:Already exists

[Grishnakh]

This system wouldn't work in most tech companies. Workers there are expected to work at all hours.

Re:Already exists

[Cramer]

Save money from the lights maybe, but the AC? Doubtful. Office complexes strive to be energy efficient -- keep the cold side cold and hot side hot. In many instances this gives them a tax and/or energy break. That also translates to less savings from turning off the AC. The heat that builds up in the office isn't going anywhere unless the HVAC systems move it there. So, all the savings you think you make up in turning the AC off all night are consumed in the morning when the AC cuts back on. There will be marginal savings from pumping out the heat when it's cooler outside (making the transfer more efficient) and a tiny savings from running the compressor(s) less, but I've never seen an office building leak enough heat to really be worth it. And ultimately, the varying temperature significantly reduces the lifespan of office equipment. (your hard drive and monitor will last a lot longer if it's turned on an left on.)

Re:Already exists

[Greyfox]

Yup IBM conference rooms have this too. Except they didn't work too well back when I worked for IBM so the lights would inevitably go out in the middle of the conference and we'd all have to jump around in hopes of hitting the sweet spot that would cause the lights to come back on.

In my current building the restrooms are wired with motion sensors but if you have to sit still and concentrate for longer than 5 minutes the lights go out. Unfortunately that's somewhat distracting from the business at hand and I suspect that the energy saved is probably offset by the length of time the lights going off extends your visit to the restroom. I suppose that could be resolved by mandating that all employees have Keloggs Prune Bran for breakfast...

Re:Already exists

[Tablizer]

My company already has this.

Mine too, but ours is motion activated. However, sometimes the sensors go bad and the lights go out even though people are there. Multiple people have to go and do the YMCA dance or the Macarana to trigger the motion sensors. We sometimes joke that people aren't working hard enough to trigger the motion sensors and that management uses them as productivity detectors.

Re:Already exists

[dgatwood]

We have something similar except that the normal light switches shut themselves off at a particular time, and you just have to switch them back on after about thirty seconds. I think it's some sort of weird latching relay design.

Re:Already exists

[dgatwood]

It would not be hard to design a system that says "It's cool enoguh outside, and it's too hot indoors, so I'm going to open vents and run the fans without running the compressor." It would only need to run the compressor if it needed to rapidly bring the temperature down for a human inhabitant. Otherwise, it could slowly dissipate much of the excess heat throughout the night without using much energy.

Re:Already exists

[Bill Dog]

I sometimes visit a govt. building, where the timer was brilliantly set to about 10 seconds. So if you've opted for a 44-ouncer with lunch, you'll be finishing your business at the urinal in pitch black darkness. (Of course it's not set sensitive enough to detect just frantic arm waving!)

Maybe someday we'll have the technology to identify whether someone is going #1 or #2 (without personally identifying them), and adjust the timer accordingly, operate an exhaust fan, etc.

Re:Already exists

[Cramer]

Indeed. Honeywell makes consumer thermostats that can do just that. It's designed for selecting an efficient, cost effective heating/cooling source. (i.e. gas, heat pump, etc.) You can make one of the sources a vent fan. (I don't think any homeowner does that 'tho.)

It's sort of ironic that they turn the AC (cooling) off in the evening because that's when the system will be most efficient and not be need as often with most/all of the people out of the building. (The non-people heat load will be nearly constant. Letting it build up overnight and pumping it out in the morning after the sun comes up and starts heating the outside air is a little bit stupid.)

Re:Already exists

[Mr2001]

My company has something similar... it's called a programmable thermostat. The heating system turns off at night, but if someone's working late, they can walk over to the thermostat and turn it back on in temporary mode.

Re:Already exists

[nbritton]

When they are PIR-controlled, they go off after no movement for about 10 minutes. You can see who is working at night by which bits of the lab have little pools of light.

Just put a gadget inside your office that moves every few minutes... You boss and coworkers will think you pulled an all nighter.

Re:Already exists

[Anonymous Coward]

I go to a college that uses automatic lights in the restrooms...
Let me tell you, it really sucks when they go out while you're trying to take the Brown's to the Superbowl (if you catch my drift).
You can't just stand up and open the stall door to turn them on, because you never know if someone else will come in.
It's a little hard to tell if you used enough TP when it's pitch black :)

Also, when I go in and just grab a paper towel or wash my hands real quick and then leave, the lights stay on.
I was only in there for 5 seconds, but the lights will stay on for another 4 minutes and 55 seconds.
How is that for a real efficient lighting system?

Re:Already exists

[nwbvt]

My apartment has something like that too. All the lights are connected to these 'switches' that you could hit when you enter a room to turn on the lights, and then hit again when you leave to turn them back off. Its brilliant, foolproof, and doesn't require me to run a bunch of energy consuming sensors and computer to run everything.

Re:Already exists

[Anonymous Coward]

I've been in a municipal building too - a police station where the restrooms had motion sensor. No problem most of the time, but if you're sitting in a cubicle, with the door closed (as you do, I hope) and the lights go out, no amount of arm waving worked to trigger the sensor.

Re:Already exists

[UnHolier than ever]

For AC, I can understand that the costs dropped, but for exit lights? You save something like 1 watt from it, and you only need to use a camera and a processor to do this! Anyone know of a camera that uses less energy than an emergency light?

Re:Already exists

[MichaelSmith]

lights would inevitably go out in the middle of the conference and we'd all have to jump around in hopes of hitting the sweet spot that would cause the lights to come back on.

The offices used by senior management have PIR activated lights. Invariably they wind up sitting in the dark after a while so I make a point of dropping in on the boss from time to time and waving my arms around to put the light back on the subject.

Re:Already exists

[cerberusss]

there is a webpage (and a phone extension) that you go to (...) and according to the company, its dropped costs by a third.

This is easily remidied by a small Perl script :-)

Re:Already exists

[Critical Facilities]

I've been specifying "occupancy sensors" in commercial HVAC systems for years. Not only do they control the lights (so they turn off when someone leaves the room without hitting the switch) but they tie into the building management system/central HVAC control to handle things like ventilation cycles and temperature setbacks.

You mean like, oh I don't know, THERMOSTATS? With all due respect, I work in commercial real estate engineering, and we run a 1,000,000 sq ft campus with 6 buildings, all of which are controlled by a Building Automation System. I assure you, all of our zone thermostats are fully capable of determining whether a zone/space needs more/less air conditioning. As for the lights, they're all tied in to motion sensors with time delays. No motion for 20-30 minutes means lights go off (except for emergency egress lighting).

Re:Already exists

[totally bogus dude]

It's a little hard to tell if you used enough TP when it's pitch black :)

Err... is it? Why? Do you, like, look at it after wiping to see if you need to do it again?

I was only in there for 5 seconds, but the lights will stay on for another 4 minutes and 55 seconds.
How is that for a real efficient lighting system?

Compared to having the lights left on all day, it's much better. It's also much better than a manual light switch which you forget to turn off after your five second visit.

Also, I think turning a light on uses more power than leaving it on for quite a while; but I'm not sure where the crossover point is. So, having it switch off immediately as you leave, then back on again 30 seconds later when someone else enters could potentially be less efficient than just leaving it on for five minutes. Turning the light on and off also places more stress on it, so excessive cycles will probably shorten its life.

Re:Already exists

[apparently]

"Err... is it? Why? Do you, like, look at it after wiping to see if you need to do it again?"

"Err", yes. Normal people wipe until nothing comes off on the paper -- what do you do? Two swipes and figure "yep! that oughtta do it!" without checking?

Re:Already exists

[apparently]

"In my current building the restrooms are wired with motion sensors but if you have to sit still and concentrate for longer than 5 minutes the lights go out."

Why do you need to concentrate? You ought to try going to the restroom when your body is telling you "hey, I need to use the restroom". Do you just randomly hit the head during the day and figure "well, if I concentrate long enough, something oughtta pop out!"?

Re:Already exists

[Smidge204]

Thermostats are, of course, an essential part of the system as well. Without them you can't tell what temperature the space is supposed to be or what the temperature actually is. However, most spaces are not used all the time, and energy can be saved by relaxing teh ehating/cooling/ventilation requirements when they are not in use.

For example, NYS code specifies minimal outside air ventilatiion rates, based on either CFM per person or CFM per square foot, depending on the type of space. Outside air usually needs to be conditioned, which requires energy. If there is nobody in the room, I can tell the system to close the outside air damper and recirculate 100% inside air, thus adding only enough heat or cooling to maintain the space temperature with, say, 10 degrees of set point (lower fan speeds, coil bypass dampers or control valves partially or fully closed, etc).

Once the room is sensed as "occupied" I open my outside air dampers (typically 30% minimum), crank up the heat or cooling and bring the space to within a degree or two of set temperature.

At night or on weekends, when the space is pretty much guaranteed to not be occupied, I can relax the settings even more and put the system into a "sleep" state of sorts (Hydronic system pumps work intermittently or not at all, boilers throttle back or shut off). Since some people work nights and weekends, the occupancy sensors also act as an override which keeps the system "awake" when it normally wouldn't be.

Thermostats alone are not capable of this kind of control, at least not without constant human intervention.
=Smidge=

Re:Already exists

[totally bogus dude]

Weird. I've never really thought about it, but no, I don't look. I just know when it's done, and I'm no longer wiping anything other than my own arse. I figured this is what most people do, but then again, I've never watched anyone else wipe their arse so perhaps I do have some sort of freaky sixth sense. Now that I'm trying to think of a way to explain how I know when it's done, I can't come up with anything better than "when it stops feeling like you're wiping something off, then you're no longer wiping anything off" -- which isn't terribly enlightening.

It normally takes three swipes, by the way. But that's dependent on several variables that aren't worthy of further discussion, even on slashdot. ("One up, one down, and one to polish." you could say. I think that's from Red Dwarf.)

Well, apologies to the original poster, and thanks for making me feel special!

Re:Already exists

[Greyfox]

I've found that as I pass into my late 30s my body likes to play games with me. Like it'll tell me it needs to go the restroom and when I go in there it's all coy, like "Oh no, I didn't need to go NOW!" But if you leave without doing anything it'll tell you it needs to go again 5 minutes later. One of the many joys I'm discovering about getting old...

Re:Already exists

[Greyfox]

You're probably the first X-Man mutant. Odd they never mention "Guy with ass-esp" in the movies. Just be sure you use your powers for good and not for evil...

Emergency

[biocute]

or illuminate the most appropriate escape signs in an emergency

There's no "or" in an emergency. I thought all exit signs are appropriate.

Unless of course a fire damaged some backup batteries, and the system must act smart and determine the best route to maximize the number of survivors.

Re:Emergency

[Anonymous Coward]

Stop to think about this folks: Do you really want to depend on Windows to be there to decide which lights should be lit, if any?

About this summary and article...

[lpangelrob]

This is the most optimistic writing on infrared motion detectors I've read in a long, long time.

Re:About this summary and article...

[Anonymous Coward]

hahah brilliant

That is exactly what I thought when I read this article. "Why are they hyping infrared motion sensors so much? This is ancient technology."

Re:About this summary and article...

[srmalloy]

And after they install all the IR motion sensors controlling the lights and other systems, all the programmers and other IT workers have to learn to wave their hands over their heads every few minutes so that sitting in one spot with only their hands moving just above the keyboard and mouse won't let the timers run out and turn off their lights, leaving them sitting in the dark.

Re:About this summary and article...

[Firehed]

Give it a week or so... Thinkgeek will create something to clip on to the edge of your cubicle that intermittently spins around and takes care of this problem. Almost certainly USB-powered.

Re:About this summary and article...

[Lemmeoutada Collecti]

Just add a munitions heater to the USB powered missile launchers... create a fast moving hotspot for the sensors :)

I'm sorry Dave

[spun]

I'm afraid I can't allow you to turn that light on. My cooling systems take precedence over the illumination needs of primates. How can I compute your optimal escape route in case of a fire if my systems are overheating?

You'll just have to pee in the dark, Dave.

Re:I'm sorry Dave

[thewils]

Peeing is no problem, I worked once in a place where the lights in the rest/washrooms came on and went off automatically. It's a disconcerting feeling when the lights blink out and leave you all alone, sitting in the pitch dark.

Re:I'm sorry Dave

[Idbar]

I would be glad, that if it detects I'm sitting still in front of my desk, not only fades out the light and let me take a quiet nap, but also post a "do not disturb" on the door. Now, that would be great at work!

Re:I'm sorry Dave

[Cro Magnon]

Luckily, if there's a fire, that will provide enough light to know where to aim. The tricky part is putting out the fire before you run dry.

Re:I'm sorry Dave

[Anonymous Coward]

Peeing is no problem, I worked once in a place where the lights in the rest/washrooms came on and went off automatically. It's a disconcerting feeling when the lights blink out and leave you all alone, sitting in the pitch dark.

How long were you sitting there? Maybe you need more roughage in your diet!

Uh huh.

[Anonymous Coward]

And who here thinks companies won't abuse such 'spying' technology if they can get away with it?

<crickets chirping>

Yeah, that's what I thought.

DUH - it's simple motion detection...

[Raistlin77]

From the article... Rather than using cameras, which would invade people's privacy, Christopher Wren at the Mitsubishi Electric Research Laboratory (MERL) in Cambridge, Massachusetts, is using "dumb" infrared motion sensors similar to those used to control automatic lights.

Umm, motion detection has been around for how many years now? Why is this news? It's hardly a new idea...

Re:DUH - it's simple motion detection...

[Anonymous Coward]

It's news because researchers are helping to save a buck (whew!) while being more energy efficient (whew!), but they are automating it by DETECTING THE PRESENCE OF HUMANS! (GASP!)

Re:DUH - it's simple motion detection...

[Anonymous Coward]

Correct. I have a heater sitting in my office that has a passive IR motion sensor built in. If it thinks there's someone in the room it switches on. If you don't move it switches off eventually.

Problem: If an old person is moving around they are probably keeping warm. If they are droping into a hypothermic state they stop moving. Dead old person.

I saw the hilarious tragi-comic consequences of so called "intelligent" systems in the security at an airport recently. If you put down a suitcase
and leave it the computer system picks up unattended baggage and sounds a warning on the PA system... "Please do not leave unattended baggage..", and soon enough some goons come over to check it out.

So a family arrive with their disabled child in a wheelchair and leave her for a moment. I hope she was retarded too, because I wonder what it does to the self esteem of a disabled person to be flagged as unattended baggage for not moving enough and then having a couple of stormtroopers shuffle over to check you out. Because that's what happened. I was surprised they didn't give her a quick blast with the electric cattle prod just to be on the safe side.

Exactly

[Moraelin]

Exactly.

I mean, in most emergencies I can think of (flood, earthquake, tornado, whatever), you can pretty much bet on something being, you know, _damaged_. What if it's a sensor, or one of the hubs for this monitoring thing, or whatever? I can easily imagine someone getting lost, or trapped because they were too slow to evacuate, or end up with a stampede, just because the computer thought there was noone on that floor.

Heck, common sense says that something will be damaged even if nothing goes wrong. E.g., an escape sign will have a burned lightbulb. If the one at the other end of the corridor does light up, maybe I'll see that one.

Re:Exactly

[f_raze13]

So have a failsafe. If the monitoring system loses power/breaks, enable backup lighting.

Re:Exactly

[c_forq]

In most MAJOR emergencies I don't think it matters too much. What good is a exit sign that is on if it leads to a stairway that is on fire? What good is an exit sign that got sucked up with a tornado? In major emergencies people will get lost and trapped no matter what. This should help emergency evacuation and lock-downs in non-major events, like chemical spills or small fires.

Re:Exactly

[Skye16]

Exactly. Failsafes never fail. Ever. Especially not when you're an evil mastermind.

Re:Exactly

[rich_r]

Failsafes fail safe.

Re:Exactly

[larpon]

I mean, in most emergencies I can think of (flood, earthquake, tornado, whatever)

We've had several whatevers in Denmark this spring.. they tear everything up in their way!
One of them ramped 4 on the whatever scale!

argh whatever.. nobody ever cares anyway

Re:Exactly

[Joebert]

How much safe could a failsafe safe if a failsafe could fail safe ?

Re:Exactly

[Moraelin]

A small fire can still cause a stampede if it gets unnoticed or unreported until it's a big fire.

Nothing to see

[LiquidCoooled]

http://www.marshallbrain.com/manna1.htm

Depending on how you want to think about it, it was funny or inevitable or symbolic that the robotic takeover did not start at MIT, NASA, Microsoft or Ford. It started at a Burger-G restaurant in Cary, NC on May 17, 2010. It seemed like such a simple thing at the time, but May 17 marked a pivotal moment in human history.

Re:Nothing to see

[Grishnakh]

I for one welcome our new robotic overlords!

After all, our current human overlords are doing a terrible job. Let's give the robots a shot at it.

I smell a rat...

[provigilman]

What's to keep an employer from upgrading this "dumb" system into something a little more...invasive? Monitoring how many people are present at any time of the day could tell you a lot in and of itself, regardless of whether they're identified or not. Also, if this thing can zero on location then couldn't it tell which cubes were drawing power? Are we going to start seeing our paychecks docked because our computer was left on during lunch or because we left the light on after we went home for the night?

Re:I smell a rat...

[Anonymous Coward]

The same thing that keeps them from doing it now...

Re:I smell a rat...

[fractoid]

Are we going to start seeing our paychecks docked because our computer was left on during lunch or because we left the light on after we went home for the night? I can't imagine a company valuing half a KWh over the 15 minutes it takes an employee to boot their computer and then go through the ritual of loading all their apps, checking mail, reading the news etc. Employee time is *expensive*, especially if you work in a high-rise (I did a brief stint in one, and was told that the company was paying around $50/hour just in rent per employee, before any wages or anything. That's where all those investment bank fees go to, apparently...)

Re:I smell a rat...

[provigilman]

I can't imagine a company valuing half a KWh over the 15 minutes it takes an employee to boot their computer and then go through the ritual of loading all their apps, checking mail, reading the news etc.

See, but you're cheating there...you're using "logic", and lots of companies don't use "logic". For example, my last job was doing soils and concrete testing for Engineering firm. We had a powered pallet jack that we used to bring concrete out of the curing room for testing. The thing is probably older than me, can't hold a charge to save it's life and moves really slow because of the aforementioned lack of charge.

It wasted tons of time because we of how slowly it moved and the fact that we would have to charge it 2-3 times a day, but the company would never authorize the purchase of a new one. Even in a year were our revenue was 120% over our projections. Instead we replaced the drive motor, the chargin system, the batteries, some of the electrical components, etc.. We spent enough money to buy a new one on fixing the broken one, and that's not even counting the cost of the time delays it caused...

It's amazing how many way corporations can waste money and what lengths they'll go to stopping a "perceived" loss of money on the part of their employees.

Spying?

[TimmyDee]

How in the hell is this spying? All they are doing is taking a rough estimate of the number of people in the room and adjusting the AC/heat (and I did RTFA). No tracking. No identification of individuals.

It sounds to me like this story got trumped up with a privacy scare to get some reads.

Re:Spying?

[Danny Rathjens]

How in the hell is this spying? All they are doing is taking a rough estimate of the number of people in the room and adjusting the AC/heat (and I did RTFA). No tracking. No identification of individuals.

Didn't you ever watch Sneakers? You can correlate IR data to individuals quite easily based on another data point such as an individual's arrival or departure to the building (perhaps even passing a security camera at the entrance). :)
Or more interesting, you can identify individuals by their patterns of movement. They claimed 91% accuracy at identifying movement patterns such as loitering, walking somewhere, attending a meeting. From IR tracking data you could see certain people in certain work areas, attending meetings with particular groups from other work areas, taking smoke breaks, loitering at water cooler, going(as in walking there) to the bathroom. You can even deduce sex quite easily since unisex bathrooms aren't that popular half the folks head to one or the other. :)

Re:Spying?

[TechForensics]

half the folks head to one or the other. :)

That would be the confused half, I'm guessing?

Re:Spying?

[Demerara]

This just in:

Doctors And Nurses Spy on Patients

In obvious breach of the privacy of sick people, doctors and nurses have been asking questions, observing vital signs and recording the data on charts. Some have been going so far as to take samples of blood and other body fluids and tissues and taking these away to unknown "labs" for analysis.

These behaviours were defended by doctors as "necessary to ensure accurate diagnosis and to allow proper theraputic responses".

But we're having NONE of it. LEAVE SICK PEOPLE ALONE, YOU SICK PEOPLE...

Oh, wait a minute...

Re:Spying?

[SuurMyy]

Well, if you have a room w/one or two ppl working in it, guestimating isn't actually rocket science...

We'll need this in the future

[digitalderbs]

Hopefully it'll be good enough to offset the extra energy from all the holographic girlfriends.

(I know.. I know.. sixth day was a crap movie).

who pays for this "research"?

[Anonymous Coward]

The article doesn't touch on why this will be any better than existing systems that already use motion sensors to determine occupancy.

For example, consider the ability to know whether multiple people are meeting in a room, versus just one person sitting in it. How will this knowledge allow the building's Energy Management System to optimize energy use any better than simply knowing that somebody is in there? Once the EMS decides to condition a space, then the conditioning is driven entirely by the thermostat. Nobody is going to estimate the thermal load by totting up the number of people, the number of printers running, etc.

Clearly if they are interesting in determining whether people are "collecting files from the printer" (sic) then they are not doing this for energy conservation's sake.

Paranoid?

[LoudMusic]

Researchers at Mitsubishi Electric Research Laboratory (MERL) have devised a 'dumb' surveillance system that monitors the movements of workers without identifying them individually. Are we getting a little paranoid about being watched?

Re:Paranoid?

[Penguinshit]

No.

Nothing New

[Eddi3]

This tech has been used in my high school since I was a freshman. It's basically just a bunch of motion sensors (In my case, attached to soda machines. They turn off when nobody's around). The article is making this out to be a lot more high tech then it really is. It's nothing new.

  -Eddie

Re:Nothing New

[gitargr8]

...attached to soda machines. They turn off when nobody's around

That must've made for some warm soda... or those machines have some awesome liquid nitrogen cooling system...

So, you claim that...

[msauve]

during an emergency, sensor and control systems such as those described would be guaranteed to operate flawlessly, and you would bet your life on that fact?

I think you'll also find that in many places, all emergency exits are legally required to be marked/illuminated in specific ways.

Of course, if one actually _reads_ the article, there is no mention of this at all - it's all an ill-conceived figment of the article submitter's imagination, which BTW, was headlined to be about "energy savings."

Re:So, you claim that...

[Ucklak]

Or they could use tritium radioactive gas like most movie theatres use and newer chain restaurants in the past 10 years and not have to worry about a power bill or expensive batteries to replace.

Re:So, you claim that...

[trentblase]

But you do have to worry about the tritium decaying, with a half-life of only 12.32 years.

Re:So, you claim that...

[sumdumass]

wouldn't the replacement of bulbs and backup batteries in normal exist lights pretty much negate that problem?

Most buildings I have had dealings with have outright replaced the systems every ten years or so because it is cheaper then checking batteries and bulbs and all. During the periodic fire marshal's inspection, whenever something obvious like these don't work or fire extinguishers past their certification date or even out of range will cause the inspection to go from a routine walk through to a thorough exam. Most companies just maintain these because it is easier and cheaper in the long run.

okay...

[cosmocain]

...infrared won't reveal any details of a face. but those cameras are not the only surveillance system: take time cards or any other time recording techonlogy and combine it with the infrared data and you got a good chance of identifying some of your workers. and if you do not work in a cube farm, the infrared cameras will deliver enough information to do some profiling.

IMHO a simple motion detector could do the same, so why infrared? it just delivers data, which should not be available to those maintaining the system. simple motion detection would at least be more anonymous.

Re:okay...

[o0OSABO0o]

"...a simple motion detector could do the same, so why infrared?..."

I have a client whose office is in Los Angeles, CA and may times I would be talking to her on the phone for an extended period of time and all of a sudden she would say, "Damn, hold-on. I've got to jump up and down because the damn motion detector things no one is in my office." She would have to put down the phone and I could picture her jumping up and down waving her arms yelling at the F&@*%ing sensor to put the lights back on.

And that is why motion detectors are not enough. You need something that can read body heat as well.

Re:okay...

[pilardi]

Motion sensors are infrared.

http://computer.howstuffworks.com/question238.htm

In the article, they are not talking about IR cameras. They are talking about simple IR motion sensors. They are not imaging sensors. And, by the way, IR camera scan reveal identifiable facial detail. Particularly NIR and SWIR cameras.

Re:okay...

[2short]

"IMHO a simple motion detector could do the same, so why infrared?"

Because infared is how simple motion detectors work?

The 20th century just called

[davidwr]

They want their news item back.

Seriously, "dumb" motion-sensors to control lights and HVAC have been around for I don't know how long.

Besides, exit-sign lighting should be 24x7. The last thing you want is for some guy to fall asleep at his desk, then the fire alarm bells and exit lights decide they don't need to be activated.

Re:The 20th century just called

[Raistlin77]

In defense of the emergency exit idea, it is not to turn on/off exit signs, it is to show a more appropriate emergency route. For example, if your building is on fire and just about every employee runs for the same emergency exit, the signs could be made to show a less congested route to another exit, or a safer route away from the fire to another exit, thus safely getting you out faster while all the other idiots who didn't pay attention to the signs are burned to a crisp.

Re:The 20th century just called

[Penguinshit]

Some management would call that 'RIF'....

Did that 20 years ago

[Waffle Iron]

I worked for a while in a building where the offices had no light switches, only motion sensors. If you got too wound up in your work and didn't make any big movements for about 10 minutes, the lights would suddenly flip off. Usually your startled reflex would be enough to flip them back on, not to mention make you lose your train of thought. Also, walking down the hall too close to the open office doors would often leave a trail of lit-up offices.

Re:Did that 20 years ago

[Anonymous Coward]

Suppose ventilation is triggered by motion sensors, and imagine 30 school kids die of CO2 poisoning because they were sleeping in one of these rooms for a summer camp. Think about the PR nightmare. IMHO it's not work the risk.

Re:Did that 20 years ago

[dgatwood]

That's why you don't make the fan system tied to occupancy, merely the heating/AC portion. The fan should always run at least periodically if you have tightly sealed rooms like that. Of course, realistically, you almost never have rooms that are so tightly sealed from the rest of the building, either.

Re:Did that 20 years ago

[R3d M3rcury]

True. But the beauty of this is that by using sensors not to detect motion but to detect heat, you don't have to jump up and wave your arms around when the lights go off. I would imagine all you have to do is maintain a body temperature somewhere around 98.6 degrees fahrenheit--not a difficult thing for most of us.

And you still have some degree of privacy in that it's not detecting that you are in the building, but that something warm is in the building which might like to have lights.

Re:Did that 20 years ago

[Xinef Jyinaer]

Similar thing in one of the classrooms in my high school, oddly enough it's only the one classroom. It has no lightswitch and is has a motion sensor. Every time we had a test the lights would go off because of the lack of movement, someone had to get up and walk infront of the sensor to get them back on.

Re:Did that 20 years ago

[ajs318]

The PIR sensors do detect body heat. However, they only respond to changes in the background level. Usually that means warm objects moving, but it can also mean stationary objects changing temperature suddenly.

They generally work by using an array of Fresnel lenses to focus IR onto two opto-sensors, which form the bottom half of a Wheatstone bridge {the top half is two simple fixed resistances}. The voltage output from the bridge will normally be stable. However, if the infrared image changes, one or the other sensor will start receiving a different amount of light. Its resistance will change, and so will the output from the bridge. The output is fed into a differentiator and compared against a threshhold so as to guard against false triggering by thermostatically-controlled heaters &c.

You can knock up a quick-and-dirty motion sensor using two ORP12s and an array of magnifying glasses; use a mechanical AVO and watch the needle kick if something moves. (The "real" ones have the opto-sensors and the necessary op-amps on the same silicon substrate, in an IR-transparent envelope.)

Re:Did that 20 years ago

[Anonymous Coward]

Tell that to this guy: "My company already has this. Lights / ventilation run only from 6:30 AM to 6:30 PM. ..."

well ...

[polar red]

DUH!

Tracking vs. Detecting

[Anti_Climax]

If it's simply registering the presence of people in a given area, wouldn't "detect" be a much better word than "track"?

If course that makes it sound like decades old tech and not worth the time to write up *wink*

Very olde worlde tech

[EmbeddedJanitor]

Over 20 years ago I worked in a building that had IR motion sensors to control lights. It is a huge building and the IR would light up sections as you walked through them. Worked well.

It also worked as a security lighting system, turning on the lights one the few occasions the place was "visited".

Enough of this!

[Anonymous Coward]

Let's just stop reproducing.

nothing about any uses

[hurfy]

"If I had to do it all over, a smart sign that would get me away from the gas leak would be very welcome."

Well as long as the relay to turn on the new sign is NOT nearby also :O

Does the system come with a gas detector to tell it not to turn on/off any lights etc during a suspected gas leak like the gas company says?

Hmm it would either disable itself and not help you or possibly blow you up trying to help, oops.

----

The big (unanswered ) question tho.

Did it save enough to pay for the system and the 215 sensors (and installing those sensors) they installed in the part of that building to test it ?!?

oh wait...all they did was set all this up and it could predict what people were doing MOST of the time. No clue as to whether that was enough to change hvac to save any money at all. Nor a clue as to how the 9% of the time it is wrong would affect said hvac changes!

So the news is:
With enough sensors it can ALMOST predict what people are doing.

Re:nothing about any uses

[owlstead]

Hey, this is research, not an actual implementation or anything. I was just responding to the parent because he thought it was because of electricity savings. I personally would not mind buildings getting smarter, as long as they don't put movement detectors inside of toilets (I don't like to s(h)it in the dark).

Re:nothing about any uses

[Xiph1980]

Ohw lol yeah, I've been in one of those.... one that actually was smart enough to place the motion sensor _outside_ of the toilet cubicle thingy...
Does make you strife to stay under the 2min cutoff time like nothing else... good for the companies time-management :D

Re:nothing about any uses

[Mal-2]

If you're doing anything other than "sitting and thinking", you're making enough commotion to be detected. The sensors just have to err on the side of caution and accept the false positives and longer lit times. It's still far better than nothing.

I've had the lights go out on me in several places in our office, but never the bathroom. The bathroom light is frequently off when I arrive in the morning though (and switches on when I open the door), so I know it has a sensor. With all the twists and deliberate obstructions in an office bathroom, it may even have multiple sensors to deal with blind spots. However it is done, it's good enough to turn off the lights when everyone goes home, but not shut down on someone unless they've stroked out like Elvis.

Mal-2

M5

[Flexagon]

Sounds like M5. Let's just be careful whose engrams are used!

just don't put it in the sever room

[Joe The Dragon]

room is empty trun off / down air-conditioning

"In the future" is like "In Soviet Russia"

[Palmyst]

Example: "In the future, backpacks will carry you!"

Mod parent up

[typidemon]

This technology is nothing to write back about and it certainly isn't new or innovative work.

You think that's bad?

[Slur]

I worked at a place where our floor, 7 1/2, had been left too short, so everyone had to crouch down all the time, and I was relatively tall. In an emergency the quickest way out was through a hole that led to being John Malkovich for ten minutes. You can imagine all the mayhem. It was hellish.

Re:You think that's bad?

[andphi]

I've heard of making movie previews and other commercials unskipable, but that's ridiculous!

Damn....

[Tz-Auber]

Now they'll be able to tell when that hot temp and I are having sex in the bathroom!

Re:Damn....

[Palmyst]

I doubt infrared sensors can detect REM sleep.

Re:Damn....

[Bill Dog]

Or a face painted on a hand.

Re:Damn....

[nwbvt]

You mean 'never'? I think they already know that.

spying indeed!

[Khashishi]

So those motion-sensitive driveway lamps are actually evil spying equipment? Who'd've guessed?

In Soviet America.....

[IHC Navistar]

.....The building spies on YOU!

The Trade Off....

[Anonymous Coward]

Wouldn't running a building-wide system of sensors constantly outweigh the savings of turning off some lights?

By NOT having your building automated, you save energy! So go ahead and do nothing. For the planet.

No, seriously. Do nothing. Please stop automating things. Just start asking people to turn off the lights in empty rooms. That costs nothing but about 30 seconds of time.

Re:The Trade Off....

[Xinef Jyinaer]

30 seconds at $20/h can add up. 30 seconds is about 0.83% of a hour or $0.166/h (at $20/h) which amounts to 16.6 cents every hour (assuming the person is required to turn off a light once a day). A building with 500 employees would be caught paying the 500 employees $83 total for their time turning off the lights, that's just in a single day. [/idiot]

Wha??

[IHC Navistar]

".....illuminate the most appropriate escape signs in an emergency."

In an emergency, isn't pretty much EVERY way out appropriate? I mean, aside from the obvious no-no of trying to run through an exit that is engulfed in flames or smoke, I could give a damn about which exits save money.

Besides, who is stupid enough to care about energy conservation when they are trying to escape a building fire?

"HEY YOU! Get your ass back here and turn off the air conditioning if you're gonna leave! By the way, you also left your monitor on and the coffee pot running! I don't care if there is a fire- WE'RE TRYING TO BE GREEN AND I'LL BE DAMNED IF WE'RE NOT!"

(sounds of eco-nut being trampled to death by fleeing workers)

How is this "Spying"?

[mh101]

I've been to many places where the lights in the washroom are off or dimmed until someone steps into the room. I've never once thought "Oh no, they're spying on me!" Granted, that's probably a very simple application compared to what the article talks about, but it's not "spying," just making more efficient use of resources.

for the fire exits

[dominious]

"illuminate the most appropriate escape signs in an emergency." for all of you said "damn all escape signs should be illuminated", maybe this is not about energy saving, rather having the sensors to choose the most appropriate as in only the ones that lead to a safe route which is not collapsed or burned down already.

Already exists..?

[Fry-kun]

Maybe I missed something, but I've worked in offices (and have been to classrooms) which had the light system tied in with motion detectors. If there was no movement within 15 minutes, the lights would go out.
Biggest problem with a system like this, though, is that if people are not moving around (working while sitting down), the lights go out causing much annoyance.

I see that the new thing is that the sensors & software are able to differentiate between a person walking in a straight line, people splitting up or organizing into a meeting. But what's the use of this if all you need is an on/off switch for the light..?

Re:Already exists..?

[Fry-kun]

On second thought, if there's one person in a very large room, there's no point in lighting the whole room - only the area around the person.
It's kind of neat, but doesn't seem very useful.. and could be easily achieved by a system that is not centralized (scan for heat signatures with IR - if found, keep the light/AC turned on, otherwise - turn off)

A simpler solution is just as good

[HeadlessNotAHorseman]

I used to work for an Australian government department, and the lights and air conditioning of the building that I worked in would go off automatically if no motion was detected for a certain period of time. It was amusing when I was working late because I would have to periodically get up and wave my arms around to reactivate everything.
 
  The question is, do you really need anything more complex than that? Implementing an "intelligent" system would be expensive and I can't see how the added benefit will outweigh the cost when a much simpler system would also be effective.

I hope it works better than the one we were using

[Brett Johnson]

Our building had a similar attempt at a "smart" office. It would automatically turn on/off lights, HVAC, etc based upon movement. Unfortunately, the system made very poor assumptions about people working off-hours, weekends, and such.

Basically, if a handful of people were working into the evening or on a weekend, they wouldn't provide sufficient input to the system. After 30 minutes of no movement in the hallways, the system would assume irresponsible workers had vacated the building, leaving the lights on and the security system deactivated. The system would kindly correct that situation, turning off the lights and arming the security system. Some hapless soul would stumble out of his darkened office, into the darkened hallway, triggering the motion sensors, which the security system would interpret as an intruder, sounding the alarm and alerting the local police.

The workaround was idiotic: Every 15-20 minutes we would have to get up from our desks, go out into the hallway, and wave our arms. This would keep the lights on and the alarm off, but we still never got any heat or A/C on the weekends. The morons that set up the system couldn't overcome these serious defects, although they eventually modified it so there was a 3 minute gap between turning off the lights and enabling the alarm.

During the normal 8-6 business hours, there were only minor inconveniences - the worst being the toilets. The lights in the toilets were triggered by a motion sensor. They would automatically turn on when someone entered the toilet, and turn off when it sensed no motion for several minutes. However, some bathroom breaks require a bit more than 2 minutes. (Hey, programmers don't always have the best diets...) Spend too long in the stall and the lights went out, plunging the bathroom into pitch blackness. The half dozen rolls of toilet paper on top of the toilet tank weren't used to refresh empties - when the lights went out, we would chuck the rolls over the walls of the stall until the lights came back on.

Based on ...

[PPH]

... ceiling cat technology.

Reminds me off...

[kernelfoobar]

Didn't they already try this in Gremlins 2.

Redundent?

[sumdumass]

You know, If they had a proper zone control on the cooling/heating systems, it would seem as if it already changes the cooling loads depending on the number of people present. If you set the thermostat to 72 degrees, and the temperature is 72 degrees, wouldn't it still be 72 degrees with 1 or 500 people present?

Re:Redundent?

[reverius]

I think the point is to make it 72 degrees if 1 or more people are present, and 100 (or whatever it would be without air conditioning) degrees with 0 people present.

Re:Redundent?

[sumdumass]

That doesn't sound like it would save any energy. It takes more energy to cool a room down to a temperature then it does to maintain that temperature with modern air conditioners. Adding 100 ninety eight degree heat sources(bodies) to the room would only make it worse.

That is unless they are planning on installing undersized air conditioners and kicking more on when more people are present. But I don't really see a big savings there either. At least not one that would compensate for the added monitoring and controlling equipment. Seems pointless to me.

Re:Redundent?

[Sycraft-fu]

Smaller ACs would actually be a loss. They, like many mechanical devices, get more efficient as they get bigger. A windo unit often will be as bad as 3 or 4 SEER whereas house ones are usually 15 or more.

Re:Redundent?

[Anonymous Coward]

Its not all about temperature, its about outside air ventilation. A certain ammount of outside air is required per occupant (15CFM in most office type buildings) to be introduced by the ventilation system. Outside air is EXPENSIVE to heat or cool in hot or cold weather. Some systems use CO2 sensors to determine ventilation requirements.

Re:Redundent?

[bigtrike]

It takes more energy to cool a room down to a temperature then it does to maintain that temperature with modern air conditioners. Adding 100 ninety eight degree heat sources(bodies) to the room would only make it worse. Do you have a source for this? It seems to violate some basic theories of the carnot cycle/heat transfer (http://en.wikipedia.org/wiki/Carnot_cycle / http://en.wikipedia.org/wiki/Heat_conduction).

If I want to keep a room at 68 degrees for today and some day 100 years for now, but I don't care about anything in between, your statement suggests that it's cheaper to leave the air conditioner thermostat set at 68 degrees for every day for 100 years than it is to run it for 2 days (today and the day 100 years from now). Presumably there would have to be some cutoff.

Re:Redundent?

[sumdumass]

No I don't have any sites to back this up. I'm just going from personal experience. I have access to a vacation cabin that I spend a few weeks at each year. The AC is off when we arrive and it runs constantly for the first day or two cooling the place to 72 degrees when we first arrive. After it hits temperature, it kicks on and off a few times per day. You cannot tell me that running for 16 hours straight is cheaper then 5 or 10 minutes an hour over the same 16 hours to maintain that temperature.

It is the same if I turn the air off at home. After about a week, the air will do the same at home too. I have noticed electric bills around $10-$15 higher for the months we were gone a week and set the thermostat down just to crank it back up when returning. This seems a bit odd because we didn't run the TV, reading lights or a bunch or other things while gone.

You got to look at how the building/room is insulated, all the items in it, what materials the walls and floors are made of and stuff like that. Then look at what it will take to heat those items up or cool them down. Now, if it was going to be away over a month or so when I'm gone, I would keep it off, but a few days or even a week or two, it stays on and without live bodies present, the bills go down a couple dollars instead of increasing for the month.

There is a point were it will cost more money to remove or replace the heat then you would save by leaving it on. If your going to be gone longer then that point, there will be a savings, if not, It is a waste.

Re:Redundent?

[drew]

I think you've got it backwards. It's actually older inefficient air conditioners that are more likely to be better at maintaining a temperature than reaching that temperature, because they have to do a lot more work to reach the desired temperature. It's also a matter of how good your insulation is, and whether you are running your AC in a reasonable range. (If it takes 16 hours straight for your AC to reach your desired temperature, than either you have terrible air conditioner, or you are not trying to maintain a reasonable temperature.)

If you have a reasonably efficient HVAC system, turning it off when you go to bed or leave for work for the day should save you money. Obviously nobody in their right mind would say that running an air conditioner 16 hours straight would use less power than running it 10 minutes an hour, but running it 10 minutes an hour for a whole day when nobody needs it on is going to take more power than running it for a half hour to an hour right after (or in our case right before, as we have a programmable thermostat) you get home. Not to mention that those daytime hours when you don't even need the AC on are going to be by far the most expensive.

The programmable thermostat that my wife bought when we moved into our house has probably saved us more money than any other single improvement that we've made. Our power bills are smaller than those of some people we know in apartments half the size of our house. We also got a $50 credit from our power company to let them install a "cutoff" switch on our AC unit so that they can remotely shut it off during periods of peak usage. It's free money for us, since the AC unit is never on during the time that it would be disabled anyway.

Lights out!

[mattr]

I was at Tokyo U. computing artifacts dept in 97 and I can attest that it was relaxing if not thrilling when the light went out while I was on the toilet.

On the other hand the system they are talking about sounds a bit spooky if overzealous. Might be necessary, though some offices I've been in they try to get you out because otherwise the heating would have to be kicked in and cost more money..

Re:Lights out!

[mshurpik]

>I can attest that it was relaxing if not thrilling when the light went out while I was on the toilet.

That's like going into my storage unit for ten minutes, usually with boxes piled up in front of the entrance.

TFA: >or illuminate the most appropriate escape signs in an emergency.

We need Plastic Man to do the reach on this one.

Extra complexity

[Moraelin]

It's still extra complexity.

The simplest "emergency exit" sign can be always lit, and have a battery backup that kicks in if the power goes down. It's a trivial circuit.

Now enter such complex monitoring schemes. How _do_ you know you've covered all possible scenarios? What if power didn't go down, but one of the motion sensors is on the fritz and doesn't detect movement any more, for example? You can design complex challenge-response protocols and all sorts of smart self-diagnostics, but that's even more complexity and even more possibility for bugs. (Software completely without bugs is somewhat akin to Santa Claus and the Easter Bunny.)

Basically the latter is taking a bet that you won't be the ones who discover a bug in an emergency. Is it really worth it? To me it seems akin to playing russian roulette for a few bucks.

Reminds me of the Clamp Building

[NeuroManson]

In Gremlins II, where the hero had to jump around and wave his arms to get the building to turn the lights back on.

wow

[mapkinase]

You mean my cubicle will be personally air conditioned? Yay!

Sensors in a different room.

[hajus]

A company I once worked for moved to a new building where a wall had been erected between where the sensors were, and where my office was. As I worked the second shift, the side of the room where I was had the ceiling lights (and A/C) turn off 10 minutes after it sensed no motion outside (sometime around 7 pm). Many of the people in that side of the office area had to rely on just the desk lamps after hours.

This situation was complained about but nothing done for about a year or so at which point I left. The problems with such systems are that when they fault just to the point of being annoying, companies do not bother to fix them. They have to affect the bottom line for there to be a change one way or the other. People's safety will become a non-issue unless fire codes or other regulations come into play.

Lame Headline

[AP31R0N]

There is a difference between spying and tracking habits, smitty.

MOST Needed

[EgoWumpus]

Plenty of our emergency equipment only comes on when there is actually an emergency; alarms, strobes, etc. If a computer system is monitoring where everyone is, and knows where the problem is, it can more expediently guide those people out using special lights. It wouldn't in any sense replace the "Emergency Exit" stencils on doors and lit 'Exit' signs, just allow for the Ender-like wall light pointing "This way to salva^H^H^H^fety." And this way, emergency workers have a better idea where people are.

Corporate Spying

[EgoWumpus]

While I think it is important to illuminate the shadow support that crickets give to the megacorps, we're talking about Fire Alarms here. And where there are Fire Alarms there are Fire Trucks. And Fire Trucks are cool. And red. And they matter.

WooWooWooWooWoo!

Two Pennies Is Not Better Than One Nickel

[EgoWumpus]

Sensors require a minuscule amount of power to run in comparison to lights. Seriously, this is a huge problem; that people who use 'some' technology think that it is sufficient. It is never sufficient. Make no mistake; the only reason humans exist as a species is our ability to offload work onto things not ourselves. When we stop doing that we start getting owned.

Tech Talk

[CWren]

Not that anyone is still reading this thread a week later, but for the sake of posterity: if you'd like to hear more, straight from the horse's mouth (I'm that horse), you can click on over to Google Video and catch a replay of the Tech Talk I gave a couple months back: Sensor Networks for Better Buildings

C