The Rain System “contains wildfires within 10 minutes of ignition”

https://www.rain.aero/

by jessenichols • 4 years ago

181 143 4 years ago

rmason• 4 years ago

Conveniently it's a single tree that catches fire with none nearby. Plus the drone is mere minutes away. What if it's densely packed brush and the drone is further away? Or a powerline touches off the fire setting hundreds of trees aflame? Or multiple lightning strikes touches off many fires in a wide area? This thing looks as if will be very expensive and work only some of the time.

shaftway2• 4 years ago

I don't think this would even be able to put out a single tree catching fire. Generally the fire is too vertical and the extinguisher would land on the ground and not be activated by the heat, so it won't go off until flaming branches fall down, at which point the fire is too large.

davidbanham• 4 years ago

You’re referring to a canopy or crown fire. It depends on conditions, of course, but it usually takes time for enough heat to build up that a fire crowns. There’s generally a period where it will burn through the lower and mid stories and it spreads.

lucasmullens• 4 years ago

Expecting something to work 100% of the time is far too high of a bar to set for fire prevention. If it helps it saves lives.

vasco• 4 years ago

That's not necessarily true. It may help and save no lives. It may not help.

lumost• 4 years ago

Or the drone's battery/electronics light on fire.... There is no good reason why the drones themselves can't be the source of the fire.

jacquesm2• 4 years ago

Exactly my worry. To get to reasonable flight times you need very energy dense batteries which are a potential source of ignition, and good luck putting that out.

Ancapistani• 4 years ago

My intuition says that in those cases, even partially extinguishing the fire would buy a bit of time, and in the meantime you’ve now got live aerial footage of the fire’s location and progress to better inform the people who are responding to it.

… in other words, putting out the fire with the UAS is a “best case”, and may happen more often than you’d think, but really it’s mostly marketing. That doesn’t mean that the system doesn’t provide real value.

landemva• 4 years ago

It will be in great publicity videos, with FEMA kicking in funds for purchase grants. Local firefighters will have fun playing with new toys funded by further deficit spending.

onion2k3• 4 years ago

This thing looks as if will be very expensive and work only some of the time.

Something that works some of the time is significantly better than nothing, and if the times it works saves millions then the price is likely worth paying.

Don't let perfect be the enemy of good.

demadog2• 4 years ago

This is version 0.1 - it will get much better. Anything you can think of, they’ve thought of and it’s on their roadmap.

sacrosancty• 4 years ago

This thing looks as if will be very expensive and work only some of the time.

Can you identify a technology that didn't have that quality initially? Even a simple thing like a pencil - what a waste of money cutting a precision tube inside a stick and somehow installing a carbon rod, which wears out so fast you can't even finish a book without having to build a whole new one. Oh, and it randomly breaks while you're using it and requires special metal sharpening tools.

If it was obviously cheap and reliable from a quick glance, it would have already been invented at a time before it was so obvious!

missedthecue• 4 years ago

Of course, this only works if you spot the small fire quickly. By the time smoke is high up in the air, the fire has likely grown larger than this drone can handle.

systemvoltage2• 4 years ago

Drones are horizontally scalable. Instead of one, 50 can go extinguish the fire.

NAR8789• 4 years ago

That sounds like a scene straight out of factorio. I'm not sure whether that's delightful or concerning.

stickfigure• 4 years ago

You're going to need way more than 50, and they'll need to coordinate dropping on a wide flame front. Also, it's not clear what they're using as a retardant - powders and gasses are almost useless in wildland fires. You need water.

That said, what would be really useful is a surveillance drone we can send out when someone phones in a smoke report. 9 out of 10 times the call is nothing, but we still have to suit up, drive to the station, get in the engine, and go investigate.

dreamcompiler• 4 years ago

You're off by a few orders of magnitude. It's more like 50,000, and that would only work for a very small fire.

Water is heavy and it takes a lot of it to counteract the heat generated by a wildland fire. Which is why water is not the primary tool for fighting wildland fires: McLeods, chainsaws, and bulldozers are. Fuel removal along a one-dimensional perimeter is much easier than bringing in massive quantities of water across a huge two-dimensional area.

buttscicles• 4 years ago

Yeah, I'd like to know how it's spotting these fires

addaon• 4 years ago

Satellite fire detection can find ignitions pretty darn quick -- much quicker than I would expect.

Also, some ignitions come from known sources (e.g. powerline downing), and monitoring and localization already exists in some cases (PG&E).

svnt3• 4 years ago

Satellites do that.

rchaves• 4 years ago

“Rain acknowledges the essential role of wildfire in healthy ecosystems. Read More.”

I was glad to see this

soperj• 4 years ago

When you read more, it's one big buuuuuuut.

ceejayoz• 4 years ago

The "read more" largely turns it into "Rain acknowledges the essential role of wildfire in healthy ecosystems, but chooses to ignore it".

hirundo• 4 years ago

The more effective rapid fire fighting becomes, perversely there is more demand to tolerate more fuel near structures because there is less expectation of losses. If that demand is met, when the rapid response fails, the fire will be more catastrophic.

It's like, having a gun for personal security can be more dangerous than going unarmed if the gun gives you unreasonable confidence.

tobyjsullivan• 4 years ago

This is great! Sounds like they are solving the problem of detecting tiny fires in vast wilderness[0].

Detection seems like 80% of the problem. Immediate response would be a huge improvement over the current strategy. That's a win even if they don't solve the drone scaling/deployment/automation problems.

[0] https://www.rain.aero/missions/software-engineer-realtime-fi...

svnt3• 4 years ago

It looks like a practical use-case for drones even if the dropping of suppressive poke-balls probably needs improvement.

Rough numbers: 3 million acres of forest in USA, one drone could ultimately cover 500 sq km (13 km radius) 6000ish drones call it 10k for redundancy

Even at 10k per drone including base station that’s roughly $100M in hardware to cover the entirety of US forestland.

PG&E is your first customer.

edit: Even if I’m low by an order of magnitude on costs, PG&E is your first customer: https://www.utilitydive.com/news/cpuc-imposes-largest-ever-p...

python999• 4 years ago

Maybe they would just monitor the corridors along transmission lines. That would be a tiny fraction of the area.

byw• 4 years ago

Honestly $100M is not bad considering a single Boeing 737 costs the same amount.

But realistically you'd want winged drones, otherwise you'd need a large number of bases (power hookups, maintenance, etc.).

ted_dunning2• 4 years ago

Yellowstone park alone is over 2 million acres. 3 million acres is a wild under-estimate.

You are nearly TWO orders of magnitude off.

realslimjd• 4 years ago

There are 750 million acres of forest in the US and each drone can cover 40 acres. That's 18,000,000 drones required.

Jabbles• 4 years ago

They aren't actually detecting the fires, according to this article linked on their press page:

Each drone would be parked inside a secure shelter and launched in response to fire-watch cameras or satellite imagery. It would navigate autonomously to the ignition site using “computer vision,” its infrared sensors feeding data to AI software to estimate the size, growth rate, and direction of the fire, before “deciding” whether to drop retardant on the fire, create a firebreak ahead of it, or summon more drones. Human operators would oversee and approve its operations.

https://www.mercurynews.com/2020/11/30/can-bay-area-startups...

quarterdime3• 4 years ago

It seems to me like they are solving the problem of detecting tiny fires in sections of 40 acres each [1]. Yosemite is about 700k acres.

[1] https://www.businessinsider.com/startups-working-to-fight-wi...

cjensen• 4 years ago

Satellites are already used to detect hotspots. The Alert camera network [1] catches a lot too. On days where fire would be dangerous, there are already planes ready to go to get the fire before it gets large.

At best, this would provide more coverage more quickly by forward deploying drones. But planes are fast enough and you only need to use a smaller plane on a small fire, and we have plenty of them. Sure seems like techies had a bright idea who either failed to do the research or failed to give up after doing the research because it makes such a great presentation for investor storytime.

[1] https://www.alertwildfire.org

gbayes• 4 years ago

I do wonder about the range given the payload is by necessity pretty heavy

Also not going to work well with lightening / thunderstorm based ignitions, or in high wind in general

motohagiography• 4 years ago

Plus, it's a complete aerial surveillance network for wildnerness. What it doesn't prevent in fires it makes up for in policing and governance. Fire marshalls and fire codes have been used for warrantless searches on activists for decades because they have fewer restrictions than police. I once met an environmental investigator in government who was regularly brought in by regional police to do drone flyovers with FLIR/IR looking for grow ops and do paralell construction when they were more illegal as well.

Wildfire suppression is a super interesting use case for drones, and one that could scale to much larger UAVs, though the business case for them is plausibly closer to the more general enforcement case.

Accujack• 4 years ago

It's a nice effort, but not terribly practical. They can't have the drones performing surveillance over wild lands for a number of reasons, not the least of which is that a crashed drone could start a fire. They also make noise that can disturb wildlife and humans enjoying the wilderness.

Plus, they're a ready made platform for dropping stuff from the drone, so immediate military and terrorist uses.

wildzzz• 4 years ago

High altitude, long endurance drones with long range IR sensors to detect flames. Use solar panels and high power density batteries and it could fly around for months. BAE has a HALE UAV that can run for a whole year at high altitudes (PHASA-35). Use a wide angle IR camera for course detection and a zoom camera for identification. Images and location telemeters down to forest ranger base stations. It wouldn't be cheap but at least officials could react appropriately before the fire gets out of control. You could still have these small water drones to keep fires in check before real firefighters can manage it on the ground. You could even cheap out a little on the year long flight UAVs and just run them continuously during the dry season.

svnt3• 4 years ago

The drones don’t fly unless there is a fire, otherwise the drones become a major fire risk, statistically.

dragonsky674• 4 years ago

base the drones on an airship that can loiter for days/weeks.

airship could also assist in surveillance.

moralestapia• 4 years ago

Hi guys/gals, (I know HN typically prefers comments with more substance but) I just want to praise what an amazing project this is and I like how this has a real potential of saving lives, property and ecosystem. I wish you are able materialize your mission!

hexnuts• 4 years ago

I can totally see this being adapted for crowd control. Gathering has exceeded X amount of allowed people in one area, deploy tear gas and countermeasures.

mabbo2• 4 years ago

The reason we have such terrible fires now is that we keep putting out fires when they're small. Nature builds up too much burnable matter over time. Small fires would burn it in small, localized areas. But if the small fires don't happen then we get the massive fires that destroy towns, cities, and cost lives.

I am not impressed by a new way to put out small fires, because that just ensures bigger fires in the long run.

Take those drones and replace the water bombs with extra fuel tanks or batteries, so they can fly longer. Have automated patrols looking for fires daily, so that we can control small forest fires carefully- while letting them burn.

lnwlebjel• 4 years ago

Alternatively, this allows you to burn when you want to, such as when conditions are favorable to maintain control of the fire. You're correct that preventing large fires is important - control burns must be part of the solution.

RC_ITR2• 4 years ago

Yeah, we should definitely put out small fires in August and let (or force) bigger ones to burn in March.

Wild how little critical thinking gets put behind these meme ideas like 'historic forest management is the only problem and we should let everything burn'.

dillondoyle2• 4 years ago

I would be interested to hear from a scientist on a couple questions I've had around this thinking:

I've read that because of this built up material AND dead trees from beetle kill, no water or snow, etc that the fires burn too hot and kill the regenerative seeds. Is this true?

If so how does that factor into the balance of that decision?

And do they take into account the carbon and pollution emitted that's a HUGE net negative? Maybe we can't ever stop them from happening in the first place perhaps it's more flatten the curve

Plus landslides and instability.

I think I've also read burn scars don't absorb and retain as much water - but I'm not as confident in that memory?

CrazyStat• 4 years ago

Maybe we can't ever stop them from happening in the first place perhaps it's more flatten the curve

We've been trying to "flatten the curve" for the past century. Turns out that was a bad idea. The only way too get back to a healthy ecosystem at this point is to allow a hundred years of excess built up fuel to burn.

California had artificially pleasant (i.e mostly wildfire-free) summers for many years. The next few decades will be paying the price for that.

sharp112• 4 years ago

Our past forest management practices are a part of it, but climate change is also a huge part: <https://www.c2es.org/content/wildfires-and-climate-change/>

lambdadmitry• 4 years ago

I don't think it meshes well with Spanish wildfire stats [1]. I can't see any serious trend there, definitely nothing like California. The main difference is that they manage their forests and do prescribed burning.

[1] https://www.statista.com/statistics/1265354/area-burned-by-w...

tshaddox• 4 years ago

I don't think a system like this is comparable to the 10 a.m. policy of attempting to totally suppress all wildfires. It still makes sense to quickly detect and suppress all unwanted wildfires. And as far as I know, the proposed solution to the 10 a.m. policy isn't to just randomly let some percentage of natural wildfires burn as much as they want, but rather to do prescribed burns (and still fight to control and extinguish unwanted wildfires).

dabinat• 4 years ago

I read about this a few months back and it seems that a key issue with controlled burns is liability. No-one wants to be responsible if the wind blows in an unexpected direction and it gets out of control. They’re scared of the variables. And of course, the less you burn, the greater the risk when you do burn, so it’s a vicious cycle.

wbl• 4 years ago

Sovereign immunity exists for this. It's a choice to let people sue for property damage due to burns.

stevage2• 4 years ago

It is also getting much harder to do controlled burns because of climate change. The ideal weather conditions are getting rarer.

dragonsky674• 4 years ago

One of the biggest problems with managing bushfire in Australia is having the fire "jump the line". Where embers fly ahead of the main fire, starting spot fires that unless dealt with just extend the reach of the fire. Currently a lot of spot fire management is attempted using helicopters with water buckets, but this is both expensive and bloody dangerous. Being able to use semi disposable drones could assist dramatically.

There are however a number of problems I can see that this approach will have to overcome.

- You need to preposition the drone in such a way that it can remain charged and also is protected from any potential bush fire before it's utilised. This could be achieved with a fire protected, solar powered shelter, but this significantly increases the cost. Maybe they could be deployed from a long duration aircraft?

- The airspace around a bush fire is a particularly hostile environment. I'm not seeing the average off the shelf drone deal with smoke, particles and embers and still function effectively. Also the sensor package will have to deal with restricted view as well as reduced GPS/Radio caused by smoke/firestorm interference.

That said, this system could be very effective in dealing with smaller locations, close to built up areas where you can target deployment for those high risk fire ban periods.

Narkov2• 4 years ago

- The airspace around a bush fire is a particularly hostile environment.

This is a good point. Bushfires are also largely caused by high winds that are unpredictable. It makes drones carrying heavy fire retarding payloads a very difficult proposition.

mabbo2• 4 years ago

Wow, that's a really great point. If we have these sorts of drones helping to contain and manage the allowed fires, in an automated fashion, we can reduce the effort needed to do so, or perhaps even control larger fires.

Maybe I was too pessimistic.

c_o_n_v_e_x• 4 years ago

The tech already exists, however the regulatory environment is not there (yet) for wide-spread automated beyond visual line of sight (BVLOS) operations.

bamboozled2• 4 years ago

The Australian bush needs fire to germinate new plants and trees too.

bayindirh• 4 years ago

One of the biggest problems with managing bushfire in Australia is having the fire "jump the line".

It's also a big problem with pine forest fires. The trees and their cones explode and when combined with the generated heat, the fire can jump over a three or four lane road in an instant. That creates spreading and accelerating fires which can go on for days and burn extremely big areas down.

We had a similar fire around a decade ago in a region that I go yearly, and it was a sad view to say at least. Kilometers of road under the shadow of trees were barren, black/brown mountains covered with ash and other unpleasant things. The region is healing by itself and help from forest services and volunteers, but it shows that keeping fires small is very hard.

coryrc• 4 years ago

We can't let them burn at any time of year, because people have moved into the forest all over and we're liable off their property burns or they die.

First step, laws made to remove any liability for fire damage outside urban areas. Make it clear fires are coming and they must have a fire-proof house. But we won't, because the rich will use the poor who can't afford to rebuild their homes as a shield against anyone being held accountable.

tlb• 4 years ago

But the answer to this isn't to just let fires start whenever and not put them out. When they start on hot, windy days they can destroy cities. The right answer is some combination of controlled burns (on cool days) and stopping random wildfires as soon as they start.

robbedpeter2• 4 years ago

Use those drones to seed locally appropriate mushroom species with spore sprays. Even better, preemptively identify piles of deadwood and prioritize larger piles. Bonus points if the mushrooms are edible. Double bonus points if they're magic.

More mushrooms means more water retention, less fuel that's less dry overall, and a niche in the ecosystem is filled , expanding activity in the adjacent niches, which in turn speeds the disappearance of biodegradable mass.

robbedpeter2• 4 years ago

To clarify, just add spores to the water before deploying, it'll still be just as good in suppressing fires, and it reduces long term buildup. Win/Win!

ajmurmann• 4 years ago

I understand that this is true for some forests but not all. Lots of California forests need to burn their underbrush and would do so without human intervention. Out gorge in Oregon is not supposed to do this though.

mschuster91• 4 years ago

The reason we have such terrible fires now is that we keep putting out fires when they're small.

No, the reason why the fires are so terrible is because we build way too fucking close to forests. When your house is a couple hundred meters away from a forest that's prone to catch fire, say goodbye to your house in case of a large forest fire.

We need to stop encroaching on nature so hard when building settlements, just like our ancestors did - they respected that nature needs distance from humans. Not just because of fires, but also because of pandemics, landslides, floods, hurricanes, storms... nature is dangerous, no matter how technologically advanced we become as a species - it seems as if humanity has gotten a bit megalomaniac in that regard, and we are all paying the price each year for that. And it's not going to get better, only worse with how climate change has increased the severity of extreme weather events.

Alternatively: make it clear to people that in the case they want to live in a fire-prone area, they have to protect (e.g. invest in deep underground water tanks for a powerful sprinkler system) and insure themselves. There will be no public investment or action in firefighting for areas designated fire-prone. Why should firefighters have to risk their lives for people who knowingly built houses in a fire-prone area?

JoeAltmaier• 4 years ago

The surveillance and response part is interesting. I can't imagine drones that size can deliver an effective response. Especially if the fire is in a tree crown (lightning strike) and not in a neat bounded campfire ring.

Still, knowing a fire has started is half the battle.

jwineinger• 4 years ago

Hopefully the exploding balls don't fling embers away to start another fire.

excalibur• 4 years ago

All I can picture is drones attacking campfires, grills, and hookahs.

ta9882• 4 years ago

If you do a campfire in a high-risk area you probably deserve your fire to be put off.

robomartin• 4 years ago

Yeah. No. As Mr. Wonderful like to say, take it behind the barn and shoot it.

Even a swarm of these cannot compare to the awesome power of small and large (Sikorsky) helicopters as well as medium to large (Boeing 707-class) tanker aircraft attacking a fire.

I have lived in fire country half my life. You NEVER get to fires when it's a convenient pile of logs in the middle of an accessible clearing that a drone can hover 20 feet above to precision drop suppressant balls. The more likely scenario is that, by the time it is detected you have to drop swimming-pool quantities of water or suppressant multiple times. And, of course, let's not even discuss a decent amount of wind and everything that goes with the situational awareness that pilots have.

hosh• 4 years ago

Neat tech, but I think prescribed burns are probably better for the general health of the forest ecology. Furthermore, prescribed burn uses a smaller tech stack (less dependencies on high tech) and is more resilient to logisitical and tech failures.

Speaking more on resiliency, in permaculture design, wildfire analysis is one of the things you do in a sector analysis when designing a site.

onychomys• 4 years ago

I do wonder how they're going to get enough coverage. There are plenty of places where a fire might start that are dozens of miles from the nearest road. You'd presumably want to deploy the base camps in some sort of grid, which would mean a whole lot of hiking things in on horses.

micro_cam2• 4 years ago

I'm very intrigued by the potential for new wildfire tech but as someone who lives in wildfire land in Montana i have my doubts about this video... detecting putting out a campfire is a pretty idealized case.

Most fires in our area seem to be started (either by lightning or an improperly extinguished campfire) and then exist as an undetectable underground smolder for quite a while until hot, dry, windy conditions emerge. Then they can blow up incredibly quickly: https://www.youtube.com/watch?v=OxcDxp07okc

The current system of lightning strike tracking, inferred overflights, lookouts and helicopter bucket/crew drops works pretty well. They literally drop a crew who can dig down and put out the smoldering parts and then drop a huge amount of water from nearby lakes on starts to get them out.

I could absolutely see drones improving the detection of smolders/smoke plumes but the method they have for putting fires out seems like it wouldn't work on buried smolders or anything that had grown too big. I wonder if they have considered something like a swarm of drones that could scoop water from smaller closer water sources and maximize the delivery rate vs the current helicopter and tanker systems.

quarterdime3• 4 years ago

You have some good points and questions here. I spent 3 summers as a wildland fire fighter and have seen lightening strikes/small fires up to massive 100k acre affairs. I will speak from that experience, but keep in mind it was over a decade ago and it's been a while since I had to think about any of this.

As you point out, a tree struck by lightening can smolder for a long time. This can go on for weeks until it explodes on a hot, dry, windy day. To put out a lightening strike you pretty much have to cut down the tree, scrape out all the smoldery bits, and mix them with dirt and water. Then you put bare hands on every square inch of the trunk and branches to make sure you didn't miss a single hot spot. The idea that a few retardant bombs will do anything strikes me as naive.

You mention abandoned or improperly extinguished campfires as well. I would love to hear how a system like this is going to confirm that nobody is in the area before basically dropping small bombs in the area.

You point out the sophisticated systems of lightening strike detection, overflights, lookouts, etc. While not perfect, these systems are pretty good. Meanwhile the Business Insider article [1] linked from Rain's website says that a dedicated surveillance drone (doesn't carry retardant) covers a mere 40 acres! You say you live in MT... how big is 40 acres relative to your nearest wild land?

You suggest drones as a method for delivery of water. Again from experience: you can slow a fire or decrease its intensity by dropping water on it. If you want to put it out, it takes boots on the ground. To extinguish a small fire (like the one in the video), you pretty much have to mix the burning bits with mineral soil and water if you're lucky. Even though we have helicopters and air tankers, we still have hot shot, helitack, and smoke jumper crews.

Finally I am also very curious whether anyone on this team has any fire experience. They're hiring, but something tells me I wouldn't be a good fit.

micro_cam2• 4 years ago

Great points especially about needing to actually cut trees. I've never been on a fire crew though friends and family members have been hot shots and tanker pilots.

I missed the 40 acres part, thats ludicrous. I live about a mile from the 3.6+ million selway bitterroot / frank church wilderness complex ... you would need a lot of drones just to cover the parts near towns.

I do think there would be some merit in a system that could detect a smoke plume quickly and slow them enough to give crews time to safely get there especially on days with multiple starts.

qayxc• 4 years ago

Meanwhile in the world of forest management: https://www.fs.usda.gov/managing-land/prescribed-fire

temp8604• 4 years ago

On a day where there is actually high fire risk (ie. high wind in hot weather) this system will be too late and not deliver enough retardant to make any difference. In those conditions you will have tens or hundreds of square meters of fire in within that time. Actual fire ignition will usually not be in the open away from other materials as shown in the example. It may be able to stop fires on days when conditions are mild enough to let it run for the ecosystem benefits.

horsawlarway• 4 years ago

I mean - it looks a LOT like it didn't even put the fire in the video out...

Look at the logs as it starts flying back up - they're clearly still burning, even after the 4 suppressants pop, and then they conveniently cut away very quickly.

dragonsky674• 4 years ago

Just wondering what the range is on those drones. If you are supposed to be covering 100's of km of bushland you will need a significant number of drones, or they will need to have high speed and long range.

Wondering if they would be better having the fire retardant dropped from some type of lighter than air craft or maybe base the drones on a lighter than air craft that can loiter in the area for a significant period, reducing the difficulty in pre placing the drones.

anonu• 4 years ago

Going to file this one in the "Amazon Drone Delivery" ideas box...

dragonsky674• 4 years ago

I wonder if the accurate deliver of fire retardant could be achieved using automated mortars or similar, maybe terminal guidance munition delivered from a high level loitering airship. Your cost may be higher, but I'm thinking that there will be a significant loss rate with drones, and you lose the cost of prepositioning.

major505• 4 years ago

The concept is interesting, but would not be easy to deploy the extinguisher agent like they where some kind of artillery shell? Use drones to patrol and find fires, and then just bombard the position from distance?

awb• 4 years ago

Not a big fan of the noise pollution. The woods are one of the last places we can go to listen to animals, the wind or silence.

I guess this is fine if it's only deployed when necessary but I definitely don't want these things buzzing around preemptively.

anitil• 4 years ago

What are those little explosive things they're dropping? Is that a standard fire fighting substance/method?

tomc1985• 4 years ago

How are they going to get this thing on-site in 10 minutes? and I have a hard time believing a brushfire in parched areas is going to stay small enough to contain in that time. Have they seen how quickly a room in a house can be completely ablaze?

eucryphia• 4 years ago

If they dropped incendiaries during planned burning in cool weather then the fuel loads wouldn’t build up to become mega-fires on hot, windy days in summer.

The Australian CSIRO developed a similar project in the 1970’s but socialists shut it down, along with forest industry, to win elections during the 1980’s

ankaAr• 4 years ago

We needed that for Corrientes.., more than 10% of the province was on fire..., Argentina, northeast..

travisgriggs• 4 years ago

What could possibly go wrong I wonder?

I can only image what happens once people realize this is a new vector for swatting. Wonder if it will get a new name.

71a54xd• 4 years ago

I'm going to be a downer here, and would bet that since this system requires lots of drones flying around to identify fires that it's more likely for one of those sensor drones to crash and cause a fire rather than actually identifying a fire small enough for a few "fire ball" retardants to extinguish an actual fire.

4oo4• 4 years ago

I really don't get why we haven't figured out how to do controlled burns properly, lots of indigenous cultures had been doing it successfully for at least hundreds of years. And they did it all without fancy technology. I think it might just come down to the arrogance of trying to exert control over a forest rather than work symbiotically with natural processes.

I never thought about the liability angle, hopefully legislation can address that to make controlled burns more viable.

8note2• 4 years ago

They didn't build immovable homes inside the places to burn

tamiral• 4 years ago

its actually really cool to see something my team and i pitched as a concept a couple years ago for a grad project now turning into reality with drones.

ta9882• 4 years ago

Is there a risk of fire associated with various drone failure modes?

natch• 4 years ago

It looks like the explosive force of the Poké balls is actually spreading embers from the fire. And as they fly through the air the dusty / gaseous suppressant will be blown off, and the hard core of the solid glowing ember will continue flying to a new landing spot where it can spread the fire anew. Oops.

slater• 4 years ago

Direct link to their video:

https://www.youtube.com/watch?v=UW_O1WKdMdY

Geee• 4 years ago

It would make more sense to have this at home. Just smaller one that puts out small fires quickly. Convince regulators that every home needs one. Profit.

jjtheblunt• 4 years ago

I wonder what historically unnatural hunting of large predators, to the point of near extinction, causes on ecosystems.

Does it alter the food chain through herbivores and, in turn, alter the vegetation pervasive, such that it's more combustable?

bryans• 4 years ago

They set fire to a pile of tinder surrounded by a rather large ring consisting of a few large branches and logs. Then during descent, not only has the fire visibly burned out on its own, but there was barely any material there to begin with. Then in the extinguishing shot, the fire is all of a sudden raging again, but with a bunch of new wood that wasn't in the original fire. Then at the end everything is completely burned and charred (but still strongly smoldering), exactly as it would look if the fire had burned itself out within its natural boundary -- not actually impeded or stopped in some way.

Being able to autonomously send a drone to a specific location is not new. Using a sensor on a drone to detect a certain thing and perform an action is not new. So the only innovation in regard to "containing wildfires within 10 minutes of ignition" that actually matters is how to handle the fire, and they didn't address that part of it at all.

Everything about this video reeks of cheap tricks from a children's magic kit, and the whole thing feels like another dime-a-dozen startup scam looking for a quick buyout before anyone realizes they've been duped.

freshglenn• 4 years ago

It seems like the extinguishing agent would likely spread embers based on how concussive it looked.

I would love to learn the user research yall did. Hit me up if you’d like to speak with a recently retired smokejumper. You’re essentially looking to take them out of business (which is fine by me) but fires are never this small or controlled. They’re usually in an area of 1-5 acres and I don’t see how this setup scales to that magnitude.

I didn’t read all of the comments but I didn’t read any that mentioned this.

ted_dunning2• 4 years ago

If you look at recent California fires, those with single initiating events occurred under very high wind conditions (> 100kph sustained) which would make drone flight impossible. Besides, such events spread to several acres within minutes of initiation. Dropping a few water balloons will do absolutely nothing.

Other major incidents has enormous numbers of initiating events. Leading up to the CNZ complex, there were over 10,000 lighting strikes in one night in the San Francisco Bay Area alone. These drones would have been completely useless here as well.

The video also claims that a single drone can handle 40 acres. There are over 800 million acres of forest in the US. Buying 20 million drones is one thing, but keeping them flying is pretty much inconceivable given how quickly most toys of this sort quit working under field conditions.

allisdust• 4 years ago

What we need is a water balloon shooting rail gun. Useful for targeting from distance and can be loaned out to cops for crowd control when not in use.

themodelplumber• 4 years ago

those with single initiating events occurred under very high wind conditions (> 100kph sustained)

Hmm, I'm not sure what you mean by "single initiating events"...is PG&E negligence or arson counted?

I can name a number of local fires here in rural NorCal that would have been benefitted by drones with this tech. In each case a minimum of 4 aircraft were dispatched after an officer visited the scene. I have to wonder how the situation might be different with a new idea like this. Would it be so bad to keep trying and improving on such concepts?

kfarr2• 4 years ago

The drones also have sizable batteries. If it crashes in high winds and then starts to burn not sure it’s being so helpful. Then who is liable for that?

jcrawfordor2• 4 years ago

It looks like they're just dropping Elide Fire Balls, an existing fire extinguishing product of limited commercial success. I'd wager due to their high cost, they run something like $100 a piece... As much as a good fire extinguisher. Of course if this worked that could be a significant cost savings compared to sending a crew, but the Fire Balls are intended for enclosed environments and require direct fire exposure to activate... I just can't see it really being a reliable way to extinguish a nascent wildfire.

The video reeks of being a demo thrown together in a weekend. The cuts to debug output are just goofy and the actual fire demo is extremely contrived. The hard part of this, in my mind, is detection... And they don't tell us really anything about how they intend to handle the complex logistics of drones canvassing a substantial forest.

epmatsw• 4 years ago

I guess you have think of the cost savings vs an actual fire too. $1.6B worth of property destroyed and a $55m cleanup project after the Boulder fires just now. That can justify a whole lot of $100 fire extinguishers even if they’re only marginally more effective (which it’s not clear that these are, just saying this is a problem you could justifiably throw some serious money at)

jcrawfordor2• 4 years ago

The comment about price was less about this application and more just to explain why Elide Fire Balls are not at all common in the real world... the company tries very hard to justify their higher cost but the advantages they point to are not especially convincing. Dropping them from above might actually be a pretty convincing use-case, but they're spherical and it takes seconds (Elide likes to say 3-5 but it seems often longer in practice) for them to melt through and activate. That means that in probably 8 cases out of 10 they'll roll away from the fire before activating, making them pretty useless. You can see in the Rain demo that they seem to have built up extra wood around the fire... I think this was to stop the balls from rolling off of the fire before activating.

In a drone system like this you're probably better off dropping the powder suppressant directly from very low. But that would require more complex custom fabrication of a delivery system. I think "Rain" just used the Fire Balls because of the simplicity of building a mechanism to release them, and not really because of any actual analysis of efficacy.

In their intended applications they're supposed to be hard-mounted where a fire is expected, or lobbed at a fire like an old-fashioned fire grenade... but the slow activation time just seems to rule that out as a good option much of the time. The only really compelling use case I see for them is mounting inside of electrical and equipment enclosures for automatic fire suppression, but there are other, cheaper products on the market for that use.

happytiger• 4 years ago

Look, the use case might be presented as ignition suppression, but I can see this being much more useful for a different use case.

I was involved with the Caldor fire near Lake Tahoe, California. That fire had a massive line put in for days, but a big wind caused multiple ignitions behind the line at night. The wind and the night prevented aircraft from flying. But drones like these could have flown. And they could have managed the tough terrain. And they could have suppressed the spot fires beyond the line, and helped that line to hold. And that would have meant potentially saving a city. Don't get me wrong, South Lake Tahoe survived due to hard work and a lot of luck, but as we know entire cities are burning and the loss of that line wiped entire neighborhoods off the map. These machines need to be scaled up to handle significant suppression requirements — something far bigger than these drones capable of helping hold a line and work in difficult conditions. I don't think small drones are it. But precision delivery, and the fact that you get immediate feedback on the effectiveness of the suppression from the drone, could be really incredible.

go_prodev• 4 years ago

Once the fire climbs a tree I think the balls will lose all effectiveness.

RickJWagner• 4 years ago

That's interesting.

But given that heat rises, I wonder how effective it would be against bigger fires. Maybe if it had a little side-launch capacity it could be even more capable?

The Rain System “contains wildfires within 10 minutes of ignition”

Related Stories