Drones and Sensors Could Spot Fires Before They Go Wild (ieee.org) 43
To detect wildfires earlier on, some researchers are proposing a novel solution that harnesses a network of Internet of Things (IoT) sensors and a fleet of drones, or unmanned aerial vehicles (UAVs). IEEE Spectrum reports: The researchers tested their approach through simulations, described in a study published May 5 in IEEE Internet of Things Journal, finding that it can detect fires that are just 2.5 km2 (just under one square mile) in size with near perfect accuracy. "In the last few years, the number, frequency, and severity of wildfires have increased dramatically worldwide, significantly impacting countries' economies, ecosystems, and communities. Wildfire management presents a significant challenge in which early fire detection is key," emphasizes Osama Bushnaq, a senior researcher at the Autonomous Robotics Research Center of the Technology Innovation Institute in Abu Dhabi, who was involved in the study.
The approach that Bushnaq and his colleagues are proposing involves a network of IoT sensors scattered throughout regions of concern, such as a national park or forests situated near communities. If a fire ignites, IoT devices deployed in the area will detect it and wait until a patrolling UAV is within transmission range to report their measurements. If a UAV receives multiple positive detections by the IoT devices, it will notify the nearby firefighting department that a wildfire has been verified.
The researchers evaluated a number of different UAVs and IoT sensors based on cost and features to determine the optimal combinations. Next, they tested their UAV-IoT approach through simulations, whereby 420 IoT sensors were deployed and 18 UAVs patrolled per square kilometer of simulated forest. The system could detect fires covering 2.5 km2 with greater than 99 percent accuracy. For smaller fires covering 0.5 km2 the approach yielded 69 percent accuracy. These results suggest that, if an optimal number of UAVs and IoT devices are present, wildfires can be detected in much shorter time than with the satellite imaging.
The approach that Bushnaq and his colleagues are proposing involves a network of IoT sensors scattered throughout regions of concern, such as a national park or forests situated near communities. If a fire ignites, IoT devices deployed in the area will detect it and wait until a patrolling UAV is within transmission range to report their measurements. If a UAV receives multiple positive detections by the IoT devices, it will notify the nearby firefighting department that a wildfire has been verified.
The researchers evaluated a number of different UAVs and IoT sensors based on cost and features to determine the optimal combinations. Next, they tested their UAV-IoT approach through simulations, whereby 420 IoT sensors were deployed and 18 UAVs patrolled per square kilometer of simulated forest. The system could detect fires covering 2.5 km2 with greater than 99 percent accuracy. For smaller fires covering 0.5 km2 the approach yielded 69 percent accuracy. These results suggest that, if an optimal number of UAVs and IoT devices are present, wildfires can be detected in much shorter time than with the satellite imaging.
Have to ask... (Score:3, Interesting)
Re: (Score:2)
You have a higher chance of killing someone with the falling battery than starting a fire. But then again this is the site of the "fuck these fancy electric cars" crowd.
Re: (Score:3)
Re: (Score:2)
Feasible? (Score:5, Interesting)
This doesn't sound very practical. It might scale to monitoring the area around picnic and campsites, but to cover just 40% of California's national parks at that density you need over 4 million sensors placed in inaccessible places, and 18000 drones in the air monitoring them. It would certainly stimulate the economy by creating jobs, but it would rather detract from the wilderness feel of the parks.
Re: Feasible? (Score:2)
In other news, deploying one human per square km (we'll call it the IoH so VC's will fund it) can also detect 100% of incipient wildfires 0.5 km^2 and larger, using their IoNaE (internet of noses and eyeballs). Because you'd need that many people constantly replacing sensors and locating lost or damaged drones.
"Daddy, why is the forest so loud?"
"Because of wildfires, son".
Re: (Score:2)
This doesn't sound very practical. It might scale to monitoring the area around picnic and campsites, but to cover just 40% of California's national parks at that density you need over 4 million sensors placed in inaccessible places, and 18000 drones in the air monitoring them. It would certainly stimulate the economy by creating jobs, but it would rather detract from the wilderness feel of the parks.
Umm, 420 IoT sensors were deployed. Let’s just say they may have been sampling the “trees” already.
Re: (Score:2)
This doesn't sound very practical. It might scale to monitoring the area around picnic and campsites,
That's good enough. We don't install motion-sensor cameras or smoke detectors every inch of our houses. We just distribute them in places where they have a good probability of detecting something before it escalates.
This is the same principle. Surely there is a cost associated with maintaining and deploying such drones, but that's just "insurance" against the horrendous costs of unchecked wildfires (in property and human lives.)
but to cover just 40% of California's national parks at that density you need over 4 million sensors placed in inaccessible places, and 18000 drones in the air monitoring them.
This is a bit of a strawman, to be honest. The researchers aren't calling to c
Re: (Score:2)
I don't know how accurate the summary is, but it does say that
so I think I was addressing a point that was raised rather than a strawman. Although I admit that 40% was a wild guess at the proportion of California's national parks which are at risk of forest fires, chosen mainly to get a nice round number of 10000 kilometres squared.
This is the way. (Score:2)
Smells like pork.
Lack of quick satellite coverage (Score:3)
I was wondering if Starlink could see forest fires from radiation but from what I can tell googling papers a forest fire does not really seem to interact with the frequencies of Starlink. NASA does have satellite observation to track wildfires though, the below link describes them. Geostationary provides coarse resolution with 5 minute updates it seems. Not an expert but the ISS is apparently equatorial and passes by San Diego (inclined) once a day. So they need something (solar powered high altitude dirigibles or gliders?).
https://www.nasa.gov/mission_p... [nasa.gov]
https://www.researchgate.net/f... [researchgate.net]
https://digital.library.adelai... [adelaide.edu.au]
Re: (Score:1)
I was wondering if Starlink could see forest fires from radiation
The top-secret optics on Elon's latest satellites can read the text on your humanities degree from LEO.
Re: (Score:2)
I was thinking along the same lines. However given the cost of wildfires in California alone, billions in real property loss, I don't know why the state wouldn't simply invest in it's own geo stationary satellite designed and dedicated to monitoring for fires.
Certainly have the expertise to do this all 'in-house', w/out pulling in the Feds. Cal-Tech, JPL, SpaceX, Vandenburg
Why the hell would you want to do this? (Score:3)
Wildfires in Alberta are spotted typically below 150 acres or about the same size that the 420 sensors and 18 UAVs get 69% of the time.
So ... massive deployment of technology and associated costs plus environmental risk of lost/damaged tech gets us to where we are at already.
I wonder how many fires would be caused by the occasional exploding IoT device battery or crashing UAV?
Re: (Score:2)
I wonder how many fires would be caused by the occasional exploding IoT device battery or crashing UAV?
Less than lightning strikes will create. Do you people think that batteries are just fire bombs waiting to explode? How on earth are you comfortable keeping a cell phone in your pocket?
Re: (Score:2)
Plus if you're doing this you're going to use the best battery technology available, and right now that is LiFePo4 which is pretty stable, it burns just like any lithium once it gets going but it's a lot harder to set it alight in the first place by abusing it.
On the other hand, I once crashed a drone into a tree and the ESC turned into fire, so the batteries are not the only hazard...
Give your head a shake (Score:2)
The claimed capability does no better than current spotting methods so why do it?
Further, they want 420 devices and 18 UAVs per km2, which in Alberta with 400,000 km2 of forest results in a deployment of 168 million IoT devices and 7.2 million UAVs. Again I emphasize, with no improvement over current methods.
Forget about battery issues; instead rephrase this as "Let's scatter 175 million pieces of electronic detritus over our bush and forest lands." You don't see any ecological issues with this?
How w
NOAA and other weather satellites (Score:2)
Re: (Score:2)
It takes about an hour to get a tanker into the air, so that thirty minutes is a big percentage of the readiness time.
Re: (Score:2)
In a previous post I thought that a dedicated satellite would be best for the state of California given the monumental costs of loss of property, lives, and supporting CalFire.
Perhaps UAV's, larger ones with capacity for water and/or retardant, strategically stationed throughout the state could be the first responders. Satellite detects a hot spot, instantly the local fire authorities, police, CalFire are notified. And one or more UAVs are deployed. If that UAV can slow the fire until the ground crews
Because nobody knows forest fires like Abu Dubai (Score:2)
FAA drone rules (Score:2)
Wondering how they are going to implement this given the draconian drone rules of the FAA? Those drones must have remote id broadcast chip and can't fly beyound visual line of sight according to current rules
Re: (Score:2)
Wrong. FAA is permitting drones for remote (out of sight) operation. They do have to broadcast their ID, but I thought that requirement wasn't active yet because the hardware doesn't exist?
Yep, it's delayed until September 16, 2023. https://www.faa.gov/uas/gettin... [faa.gov]
Regardless, that wouldn't be a show stopper exactly, if they're not picky about how you broadcast you could use BLE (which has a true broadcast mode.)
Already done from Orbit (Score:4, Informative)
I don’t know why you’d do this from the air, when it’s already done from orbit. I was involved with a wildfire fighting effort in 2015, and the data available was a mix of satellite.
As it turns out, the same type of sensors used to detect rocket launches (part of the nuclear deterrence system) is also very good at detecting wildfire.
Re: (Score:2)
Interesting AND Informative.
*invisible mod point for you.
Re: (Score:1)
Oh yes, please! (Score:3)
Wild drones... (Score:2)
Re: (Score:2)
Wild girls are piloting them. Buy the video while you still can.
Smarter drones? (Score:2)
Why not use a solar powered glider fitted with a simple optical smoke detector, GPS, and a camera. When it detects smoke particles above a certain density it takes photos of the area, tags them with a GPS location and flies home. A 10 acre blaze will generate enough smoke to be detectable over 1000 meters above ground. No need to deploy hundreds of IoT sensors when a single mobile sensor can cover large areas.
There will be some blazes started by lightning when it's not possible to to fly gliders, but the m
Re: (Score:2)
... or why do it at all when current fire spotting is better than the proposal and doesn't involve any ridiculous tech deployment?
Go talk to actual firefighters (Score:3)
2.5km2 is not a small fire. In fact, ALL of the fires involved in 2020's post Labor-day inferno were easily detected when they were much smaller than that.
Ideally you want a system that can detect fire starts when three dudes in a brush truck can easily put them out.
In practice, I think the sweet spot for IoT sensors in this problem is to provide actionable warning to homes and communities of an approaching wildfire, especially at night. And also detecting human-caused fires along roads and power lines -- and detecting such fires when they are extremely small.
Fun fact: nearly all catastrophic wildfire events are correlated with a high wind event. Wildfires not backed by strong winds don't move that fast and are generally easily avoided by humans, and rarely burn hot enough to threaten structures.
Re: (Score:2)
Ideally you want a system that can detect fire starts when three dudes in a brush truck can easily put them out.
It doesn't always work that way.
a) Sometimes the best option is to let it burn. Many forests, especially those in the west, are fire adapted ecosystems and depend on fire for their health. The fact that we've fought the fires for so long has been a huge problem.
b) Even if you do catch the fire early, that doesn't mean you can do anything about it. A fire that I was involved with in 2015 was spotted from the moment the lightning strike hit the earth. They had a fast rappel group on it same day, and even then
Re: (Score:2)
For point a.) I would agree except times of red flag (high wind, etc) danger. That's not to time to let nature take its course. So presumably someone could make an informed decision on the let it burn choice given current and predicted weather and locations at/near people.
But in practicality no bureaucrat in their right mind would ok a natural fire to burn, because were it to spread to structures or get too big, they would get blamed.
So really we are talking about a lot more prescribed burning, and less
Re: (Score:2)
At least in the case of "our" fire, the decision after the initial failure to contain it was to "let it burn" and just keep an eye on it. For about a month, it just munched away in steep terrain, slowly growing. In the mean time, there was action to reinforce firebreaks, and we were able to complete our defenses. It basically just simmered away until we wound up with a red-flag night, with > 95F temperatures, single digit humidity, and strong winds. Then it exploded.
Fortunately, for our site, the defense
Re: Go talk to actual firefighters (Score:3)
Wierdly Impractical (Score:2)
Re: Wierdly Impractical (Score:2)
A square mile? (Score:2)
Drones are dumb, use aerostats (Score:2)