Future Firefighters May Be Guided By "Robots On Reins"

Zothecula writes When firefighters need to enter smoke-filled buildings to conduct search or rescue, they frequently suffer from low visibility and often need to feel their way along walls or follow ropes reeled out by the lead firefighter. With a limited supply of oxygen carried by each firefighter, being slowed by the inability to see can severely limit their capacity to carry out duties in these environments. Now researchers from King's College London and Sheffield Hallam University have developed a prototype robot assistant for firefighters that can help guide them through even the thickest smoke.

Not so useful

[NotDrWho]

Current robots have a hard time navigating even the most basic terrain. It seems highly unlikely to me that a ground-based robot would be of any use in a burning building full of completely unpredictable and changing debris, tight spaces, random layout, etc. I doubt the prototype shown could even climb a stair, much less climb over debris.

A very small "robot" controlled by a human (aka a glorified RC helicopter) may be of some use in surveying the situation, though.

Re:

[TWX]

Boston Dynamics has been going a long way toward improving how robots handle terrain. I wouldn't be surprised if they do manage to make robots that can handle it in the near future. After all, urban battlefield conditions are probably a good analog for the chaos in a structure fire.

Re:

[NotDrWho]

Yep, and most still suck at basic walking. Even the Cujo requires a huge onboard computer just to walk. And good luck using it in a cramped space or trying to get it up a narrow set of stairs.

This is silly

[muhula]

They don't need a guide dog... too slow, hard to maintain, and move around. An infrared HUD would be perfect for this

Re:

[pixelpusher220]

My thoughts exactly. Pairing an Oculus Rift type HUD display on their already worn facemask with an infrared camera (or any other enhanced visual system) seems like a more promising route and a lot less complicated than a semi-independant robot.

Re:This is silly

[Jeff Flanagan]

This seems obvious. If it's possible for the robot to have sensors that can see in the smoke, giving the fire-fighter the sensor instead of adding the extra complexity of a guide-robot is an obviously better solution.

Re:

[RedShoeRider]

It's a good idea. It's been tried, though: http://www.flir.com/legacy/vie... [flir.com]

One of the problems with this idea is that it gives rise to the "superman" complex. Namely, that the wearer would charge into a zero visibility situation and loose situational awareness. When the unit failed/went dead/malfunctioned/leaked/whatever, you were thoroughly screwed, as it was like being plunged into a world of black ink.

I say "when" for unit failure because it really is a matter of when. Electronics exposed to the

Re:

[pixelpusher220]

You're link is something self contained 'outside' the helmet. What I'm suggesting is something inside where conditions are MUCH better for electronics.

And the robot still has the same problem with 'outside' environment, plus having to be dextrous enough to move through debris and failure leaving teams stranded. Way more complexity in that than fitting a small HUD into a helmet.

Re:

[monkeyxpress]

It looks like the main focus of the research is really around haptics. They seem to have a pretty cool whole-arm haptic interface and talk about using it to sense body motion to infer subtle control inputs from the user. That is pretty cool. They also mention that the robot can 'feel' items around it and, I am assuming, feed this back to the operator using the haptic actuators in a natural way.

I would imagine the firefighter thing is just an application domain to give them a framework for developing the tec

Re:This is silly

[CFD339]

Thermal imaging cameras are quite common for fire crews. We have one on most of our engines, trucks, and heavy rescue units. They are available built into the SCBA masks but they are extremely expensive in that configuration and can't be easily shared between crews and crew members so you'd have to buy one for each firefighter rather than one for each crew. Remember, you can't just use an IR camera from best buy. They have to be waterproof, intrinsically safe (no sparking internally when switched on/off), impact resistant, heat resistant, and 100% reliable.

Re:This is silly

[Anonymous Coward]

Speaking as a firefighter, the current generation of LWIR cameras offer excellent vision in smoke, durability, and good battery life. They are, however, not hands free so a HUD is an interesting idea, but keep in mind anything I take with me has to survive and perform in excess of 600F. Additionally, in the quest for FOV, current cameras have some fisheye distortion that makes it difficult to judge distance - firefighters have learned to live with that trade as best we can.

Re: This is silly

[ModernGeek]

I'm a firefighter, we already have infrared heads up displays. If the smoke is that thick, some windows are going to have to be broken or a hole cut in the roof so that the room can be ventilated. This adds oxygen to the fire, but it allows us to see the fire so that you can put it out. If the room is really really hot, then people don't need to be in there; and at that point you are probably only rescuing a body. Early detection and fast response are a key to controlling a fire.

Already been done

[SuperKendall]

Get yer robot firefighter here [wp.com]

Question

[Prune]

Wouldn't the resources expended in this "research" have been far better used in creating sensors for the firefighters themselves (say, a high resolution combination sonar and infrared visor) than building an awkward haptic extension with the clumsiness and poor maneuverability of current state of the art robotics? I know robots are becoming trendy, especially in the tech media, to the point of fetishism, but I sure hope no tax-backed grants supported this project, one that is ultimately a substitution of Go

Re:

[gstoddart]

You know, I should think the more help you can give these guys the better.

Because if it can lead the firefighters in, it could also have applicability in guiding victims out while leaving firefighters with free hands to watch out for other hazards as they follow the people out.

I should think most of us should STFU about what firefighters need and don't need -- if someone who runs into burning buildings says this could help save lives, I'm sure as hell not going to arm-chair quarterback that.

What's wrong wit

Re:

[CFD339]

Thank you. I commented at length about this already -- but yeah, this kind of robot would be very little help in any configuration I can currently imagine. Thermal imagine is commonly used, but handheld units are more common than the very expensive ones built into SCBA masks. You can't share a mask between crew members so you'd have to be one per firefighter rather than one per crew, and they're not cheap. Most departments have a hard enough time getting budget to replace worn out hose lines, let alon

What about Arsonist Hackers?

[Needs2BeSaid]

Great. Arsonist hackers will start fires and then guide the Firefighters into danger.

As a firefighter, I am extremely skeptical.

[CFD339]

The article gets right that visibility is limited. Let me be more clear. Visibility is often zero -- as in you may as well close your eyes. If you get there before the engine crew is actually putting water on the fire and the smoke layering is still undisturbed, you may have visibility at floor level, but often not. The minute water gets to the fire -- or a heating pipe solder joint melts and the pipe sprays water -- the building fills with steam and the layering is disturbed and conditions are zero visibility. It's also very loud, between the sounds of the fire and the sound of your breathing through the respirator mask.

Now, imagine what's on your living room floor, or your kids rooms, or blocking your hallways. Imagine you don't know the layout of your house and you're blindfolded. Try searching under those conditions, keeping in mind that seconds count as your knees are sticking to melted plastic toys and you're feeling ahead of you to make sure there's no open hole in the floor, stairway, or other hazard, and you're checking to make sure that the engineered joists holding the floor you're crawling across haven't become weakened by the heat to the point where you'll fall through into a burning basement. While doing all that, you've got one hand on the person's gear leg ahead of you (or perhaps a hose line being led by someone ahead you can't see) and your other hand is trying to sweep the floor around you with your tool, and a third hand may be trying to look around with a thermal imaging camera to find a patient on the floor, under a bed, or in a closet. You've got 20 minutes to find what you need before your low-air alarm starts going off and you've got to head out with your crew while another comes in. Meanwhile other crews are banging around trying to put the fire out before the house comes down around you.

Call me skeptical, but I don't see any current robot technology that can do all those things -- let alone do it in several hundred degree heat.

Re:

[PPalmgren]

Just wanted to say I greatly appreciate the excellent description you gave. It made me feel like I could see and feel what you experienced. Hell of a tough job, and thanks for putting in your work for all of us.

Re:

[CFD339]

Thanks for the nice feedback. I could get a lot more graphic, but I don't think people would really believe me. What you see on TV of firefighting is as unrealistic as everything else on TV. For good reason, I guess, since video of what goes on inside a real burning building would be very hard to watch. It would mostly be a dark or a white screen and a lot of noise. Not great television.

Re:

[CFD339]

SCOTT offers a mask with a heads up TIC, and I'd imagine MSA does but I've never checked. The problem is, they're expensive. We can have one handheld TIC on each engine, ladder, and heavy rescue. If we were buying the masks with built in TIC units, we'd need to have one for every pack because they really can't be shared.