CrowdSourcing SAR Searches

One of the local SAR teams, about 40 miles west of us, is conducting an ongoing search for a missing hiker. So far, the the search has come up empty handed after several attempts. And it is still uncertain whether this person even went hiking in the area that he previously said he had planned on.

Now the family of the missing hiker has funded further searching and the local SAR team with the assistance of a Discovery Channel producer and camera tech, along with a helicopter company have taken hi res photos and video of the area. These videos have started to be posted on YouTube and they are asking the public to review and see if they can spot any anomalies then email the SAR team with details.

In reviewing the first video, it is real difficult to discern and spot anything due to the terrain, weather etc. There have been some YouTube comments but so far, nothing concrete.

I am not sure though if I am in agreement with such crowdsourcing of SAR searches as it reminds me of the search for Steve Fossett. In the documentary "The Last Flight of Steve Fossett." I recall a comment to the affect that there 1000's of phone calls to the CAP and they had to change their phone number.

I suspect that with people "seeing" things that are not there in the video will result in a ton of emails and phone calls and all to no avail.

I dare say you are absolutely correct, but you reach a point in search operations where desperation prevails. Pre internet, we would reach a point in an ineffectual search where publicity was pervasive enough that we would start to get calls from psychics, who would have feelings or premonitions or dreams etc., about where the individual might be.

I remember a personal interaction with one such an individual. We had been looking for more than two weeks, finding zip. She had a vision about the location of the individual. We took her car, using her gas, and visited the area (a place we had flogged at least twice), finding nothing. It wasn't a scam. She wasn't asking for anything, but simply wanted to be at peace.

It always intrigued me that psychics always contacted us after media had elevated the operation in public awareness. As far as I know, we never received a call out of the blue, saying "I just got a message - someone is in deep doodoo in Sabino Canyon. Could you check it out?"

Yes, I am a bit skeptical, but you reach a point where you are out of options. Let the crowd sourcing begin!

Ain't the internet just wonnderful?

CrowdSourcing is an interesting term and implies that a lot of people looking at mediocre resolution video or satellite imagery will make the resolution better. Knowing where the Fossett crash site was located, did anyone go back and look at the satellite imagery on google earth to see if the aircraft could be detected? In the case of a lost hiker, what do people expect to see in G.E. resolution imagery? It's digitized imagery with definite limitations and not a high resolution print.

OT: Anyone else here shoot Kodak Panatomic-X BW film back before digital cameras replaced those archaic film cameras? That was high resolution film.

Why yes, I remember Pan-X, and that was just the stuff that was readily available. A neighbor was a lt col in the air Force, in charge of maintenance of the local u-2 squadron. He offered to arrange an overflight of our excavation, averring that they could tell whether we were using square point or round point shovels in our dig. I don't know what film they were using,but it probably had superior resolution to Pan-X. This was all part of their training routine; they had already photographed all the golf courses in Arizona, so why not do a cliff dwelling?

I would hope that beside the crowd sourcing, someone in that search effort is involving the military. We did that on the operation i referred to previously, and something flew down from Scott AFB and zapped our search area. They were not at liberty to discuss the technology involved.

I'll bet that experienced photo interpreters will be more productive than the public at large....

I participated in the GE search for Fossett.

I went back and looked at the site on GE and there was no possible way the GE search was going to work.

The resolution wasn't there and the wreckage was only a burned skeleton of fuselage rather than intact wingtips. Pictures of the wreck site (taken on foot) suggest the fuselage tubing would look like tree branches from just a few meters away.

Some folks alerted on totally routine things (like blue tarps put up by outfitters) and had to be talked down.

Without the resolution the extra eyes have no value. But if there is high resolution but not enough educated eyes to scan it all - then the crowd starts to offer a chance. But the Fossett search also revealed the lack of skill in photo interpretation and the energy required to deal with the false alerts.

unimogbert -- Thanks, that's what I figured.

Hikermor -- Pan-X was the best I found available for we mere mortals. The photo guy in my first squadron used BW high contrast copy film and a tricked out Nikon F to get some amazingly high resolution prints. I couldn't do that with a standard Nikon FM and Pan-X, but he used that high contrast film and his own dark room (he had to finesse the developing process). Is there a useful conversion to compare digital megapixels with film ASA? Or is that an apples and oranges thing?

I would think that grain size must relate somehow to pixel somehow, but I have no idea of what the conversion would be...

Google Earth resolution depends on the source. If the light is hitting it right, you can even see some of the power lines in the current image they use of my house. Older images range from off color & slightly fuzzy to very fuzzy.

I wonder if there are Probabilities of Detection figured out for this sort of Search Tactic?

Probably not enough data yet. Frankly, I think it is more of a PR device than an effective search tool. It would be different if you had a system where you relay real time, high resolution video to several trained, rested,alert observers sitting in a comfortable environment.

GE uses a wide variety of imagery from many sources. The higher resolution stuff is mostly conventional air photography from a wide variety of public domain sources (USGS, states, municipalities, etc). The satellite imagery is mostly SPOT, with about 2.5 meter resolution. The Wikipedia article on Google Earth has good info.

Since GE images will generally date from before a person became lost, they will be useless for spotting a lost person directly. However GE images can be very useful as a look at terrain and man made features for search planning. It is often used for that purpose.

To try to use imagery for spotting a lost person by "crowd sourcing" one would need high resolution images acquired after the person became lost. Apparently this was what was tried in the search mentioned in the OP.

Even then they would need a valid start point. They are essentially trying to track this person via satellite imagery without a ground truth. What did the terrain features look like just before he passed by and what changed? What would a tracker on the ground expect to see while following someone not trying to hide? A person is not going to change terrain features enough to be detected by mediocre resolution imagery without doing something big and deliberate. It won't see footprints and certainly can't determine boot-size or whether that's the right pair of Danners.

This is how desperation manifests and is another reason to stop, set up camp and get a good smokey fire going. Or carry a PLB and a signal mirror. All of the above?

Coincidentally, while enjoying some post Xmas relaxation and web surfing at my in-laws, I just ran across this: Two Views to Earth . Apparently, near realtime 1 meter resolution satellite imagery will soon be available.

The missing hiker is thought to started in the Lynn Headwaters Park area. I know that area quite well and in the lower elevation areas, it is a very popular hiking area and has well marked trails. However once you get into the back country past Norvan Falls, the terrain can be quite challenging and you need to be well equipped to attempt to hike up in there. I don't think this person would of hiked that high up into the back country, but you never know.

One of the good/bad problems this year is the low snowfall but along with the rain, cooler temperatures and short daylight hours (it gets dark up there early) means that ill prepared people will push themselves past their normal limits and find themselves in trouble before they know it. Today, it is foggy here and I would suspect that same area is also socked in with fog and clouds.

This could drastically reduce the time required for some searches, especially if the weather is good and the person who needs rescue is able to make a large, high-contrast signal, such as a big "X".

During the Fossett search there were frequent efforts to offer updated imagery because of the obvious condition that your pictures have to be taken AFTER the object has been lost.

In the case of resolution- areas I occasionally look at for hiking have some newer, much higher resolution, imagery on GE right next to areas with older, lower resolution imagery. So GE results can vary.

Near realtime imagery is expensive.

Unless you're looking at some of the street view imagery, about the best resolution I've seen on GE is 1 meter. 1 m resolution is fine for some applications, but I have my doubts that this is one of them.

One meter resoulution might help in a few SAR cases, such as when looking for a lost vehicle or downed aircraft. Looking for lost hikers, not so much. (Unless, as chaosmagnet says, they make a really big symbol.)

Note that any quoted figure on "resolution" (such as 'one meter') probably refers to optimal conditions. As chaosmagnet also implies, weather is always the bane of satellite visual imagery. Any image taken from space will always have to penetrate the entire thickness of atmosphere. Any clouds, smoke, haze, etc will degrade the quality of the image. Aircraft imagery will always have an advantage in this regard. Besides being much closer to the target (hence not requireing nearly as much optical magnification), an aircraft will be below at least part of the atmosphere. I was reminded of this while landing in Anchorage yesterday, after our Christmas visit down south. As we approached Anchorage at altitude, all one could see was the tops of a few mountains poking up through the clouds. Once we descended below the clouds on our approach, it was quite clear.

The great advantage of satellite imagery (for civilian use) is being able to see large areas at one time. Near realtime satellite views with 1 m resolution could be extremely valuable for disaster management. Think of something like a flood, hurricane, tsunami, or earthquake. An emergency response planner could immediately see which areas were hardest hit, to help decide where to deploy assets. He/she could see which bridges were still standing, and which roads were blocked. This would be very useful for routing convoys of relief supplies.

Air photo missions have been launched especially for SAR. I don't have a link handy, but I recall hearing of one a few years ago looking for missing climbers in the Alaska Range. My recollection is that they never found the missing party they were looking for, but did spot the body of a Japanese climber who had gone missing some years before. Just a bit of red parka was visible in the ice on a wall.

Now there is a great application for drones. The wonder of it is, you could unleash literally hundreds of drones to search an area. The tech we have access to nowadays will find people who are well hidden in desolate areas under adverse conditions.

I can't think of a better use for these things than SAR. Risk mitigation with drones is also a major factor. In fact, I could see where a drone could be used to locate a victim, provide valuable resources, communications etc, and give SAR teams a hard target to home in on and effect rescue.

As a public investment, especially up here in AK, I would think Drones would the real deal. Imagine what just 50 drones equipped with good AI and a full sensor array (flir, night vision, high res, sound sensing, chem trail tracking,seismic, etc) with a 1,500 mile range and a cruising speed of 250 knots could accomplish.

We had robotic ground rovers years ago that could track rabbits out in the desert. Every once in a while one would traipse through a contaminated area, then spread radioactive material into a sensitive area. They tracked the animals the same way a hound does. Pretty darned effective.

Another one of our local SAR teams did some testing with small drones earlier this fall. One advantage I see with these sized drones in our mountainous terrain which is punctuated with deep narrow ravines and gullies along with dense tree cover, is the ability to maneuver into these type of areas much easier then larger drones.

I've heard that drones are being seriously looked at for SAR applications in Alaska. I don't know any details, but I expect it is only a matter of time before we see this happening.

The USCG icebreaker Healy has experimented with using drones in addition to it's helicopter to find the best route through sea ice. The oil companies have also been experimenting with helicopter drones on the North Slope to do inspections of equipment that is difficult or dangerous to approach. Lots of potential uses besides military ones.

This might be a good time to post a reminder here.
If there is an ongoing search, and they are short of search team members, please consider posting here. Maybe with a title like "Ongoing Search at XXX Needs Volunteers". It is quite possible that members of this forum might be able to assist. The OP needs to include specific details of the exact region where the problem is occurring, and contact phones would be excellent.

I can't do this one because I'm away on holidays. But I will keep it in mind in 2014, esp. if the region is in the SouthWest.

Personally, I am a little bit skeptical about the "crowd sourcing" thing. But open sourcing of these reaches to people with a background in wilderness survival is a whole different ballgame.

thanks,
Pete

Pete, while your intentions are good, for several reasons this doesn't always work out so well in practice.

There are some issues with using "walk ons" in searches. One big problem is sorting out (in a short time frame) those folks who really have some backcountry savvy and survival skills from those who only think they do. Without having spent time with a person in the field it is tough to get a good sense of their experience. A searcher who themselves become lost, hypthermic, injured, or exhausted is no longer part of the solution, but rather they have added to the problem.

Another problem is that even if people are capable of taking care of themselves in the backcountry, they may not understand basic search methods. For example, untrained searchers can misss important clues. Or they may inadvertantly obliterate those clues. Search teams train their members to not look for people, but rather to look for clues. The idea is that if you are looking for people, you might easily miss seeing a subtle clue, but if you are looking for clues you are unlikely to miss seeing a person. Trained searchers are also taught how to document and preserve clues. Those clues can be crucial to narrowing down the search area.

Yet another issue is that untrained people probably aren't accustomed to working in ICS, and may not understand the importance of following the instructions from the search incident management team, and reporting accurately what they did. This can make it difficult to systematically cover the search area. An example from a recent search illustrates this. A bunch of local folks who were untrained but knew the area well were helping. The Ops Section Chief gave them instruction as to which area to search. But the locals had their own ideas about where the lost person was likely to be, and deviated from their assigned area. Unknown to them, that area had already been thoroughly searched. Thus one area was searched multiple times, and Ops had to send yet another team to cover the unsearched area.

However, we can sometimes put walk on volunteers to good use. One example I can think of was an avalanche body recovery (a deeply buried snow machiner who wasn't wearing a beacon). It was a big slide and turned into a major probing operation lasting a couple of days. We checked to make sure they all had appropriate gear, and gave them a quick lesson in how a probe lines works. We then grouped them into teams with an experienced person in charge of each team.

I'm afraid I must second AKSAR's extremely cogent comments. That doesn't mean that walk ons can't be beneficial, but expect to serve in a supporting role.

SAR is definitely a team activity and it takes a while to develop teamwork, especially among strangers.

My introduction to SAR was an extensive search in extremely severe winter conditions. The operation was a hideous boondoggle from beginning to end, primarily because no one, from the sheriff on down, really knew what to do, or how to do it. About the best that could be said is that the repercussions from that event lead to steadily more competent SAR in the Tucson area. And that required a lot of team building, a prerequisite for operating in critical situations.

I completely agree with both AKSAR and Hikermor.

Although I am not part of any SAR. Over the years, both my SO and I have helped and participated with SAR searches when they put the word out for help...which is not often around here as we have about 10 SAR teams in the area which provide mutual aid for each other.

All the times we have been out, after signing forms and then questioned and scrutinized on our skill levels and gear, we have been paired with SAR team members. On any search we follow their rules and protocol and stick with them at all times. Meaning no wandering off and looking on your own or thinking you know more then the SAR team. And yes, that does happen as AKSAR mentioned in his post.

If you ever decide to get involved with these searches, you must be in much better then average physical hiking shape as you can be out for hours in the hot summer sun to rain and cold, snow conditions on all kinds of different terrain. Also the requisite gear is a must and SAR ICS will check that you do have that gear before they let you out in the field.

good points made ... but I see no problem with them.
pairing people up is probably a good idea.
the main problem appears to be the "conventional wisdom" that adding extra people as the search progresses is not worth the problems and difficulties. I can see why SAR teams have come to believe this. But by the same token, there are ways that things could be kept flexible if good comms are maintained with team members (incl. regular comms checks), and someone updates the overall navigational map (who-is-where). GPS systems and mobile radios have made this more practical.

we're looking at moving into an era of decreasing resources and budgets. creative solutions are needed, or lower success rates will need to be accepted. just my $0.02.

Pete

Now there's something I would be all too happy to help develop. Wonder if anyone's out there I could contract with.

I dare say you are correct in these statements. It seems to me that they make an excellent argument for affiliating with your local SAR organization before there is an emergency. Develop the relationships, build the team, and acquire capability.

Also, while it is true that exceptional physical stamina is highly desirable, it is not necessarily a mandatory prerequisite. Our organization has people who were not up to the more strenuous tasks and all of us were careful to make appropriate mission assignments. "A man has got to know his limitations" after all. I was impressed many times with the ability of a highly disparate group of people,with many varying capabilities, to come together and work productively toward a common, worthwhile goal.

Pete, you will also get "lower success rates" if the searchers don't know what they are doing.

Hikermor nailed it. If you are really interested, then hook up with a local team. Get some training, and let people get to know you and your capabilities. Most teams are always looking for good people who want to join.

And as Teslinhiker noted, teams usually have mutual aid arrangements (formal or informal) to help out other teams in the region.

Searching is a surprisingly subtle activity. As AKSAR put it,you are looking for clues,not just human size objects, and these are often not very obvious. A competent searcher must be focused and aware. It doesn't help if you are dehydrated, sore, distracted, etc. You will miss important stuff. This has been demonstrated way too many times.

when I am searching (or rather, when I was searching), I could always tell when I was alert and focused by the fact that I was also noting archaeological material. Archaeology is my profession and I have done a lot of archaeological survey - looking for traces of human activity in the landscape. This is an activity that is obviously closely related to SAR, and the same constraints apply. It is simply amazing what people can miss if they are distracted, disoriented, or confused.

Searching and SAR work are like most human activities. You get better with practice and experience, especially if guided by decent training. If your garden needed weeding, would you crowd source that, or turn to an experienced gardener?

"Lots of potential uses besides military ones." (AKSAR)

This is certainly an emerging technology of immense promise and problems in everyday life. I am intrigued by the possibilities for archaeological and paleontological work. Of course we have been using drones, of a slightly different type, in civilian life for years.

I really wish you wouldn't call that to people's attention.
I'm somewhat embarrassed at what I've become.....and my co-workers seem to be completely unaware.....


The proliferation of UAVs may well be the death of civilian aviation. Lightplanes can't take bird strikes very well so a 4lb battery pack thru the windshield will be even worse.

Don't worry about it, Amazon will carry plenty of insurance...

From this morning's Anchorage Daily News: Expanded drone testing for commercial use another Alaska aviation first

Ben, looks like University of Alaska at Fairbanks is who you want to talk to:
Hikermor, regarding your comments:

Regarding safety, unimogbert:

The next few years should be very interesting.

"It would be different if you had a system where you relay real time, high resolution video to several trained, rested,alert observers sitting in a comfortable environment."

that may be the breakthrough. if we could get good drones with HD video over the rescue area, it could dramatically improve the search coverage. BUT it would need good observers to sit in a comfortable location and scan the HD video with careful eyes. There is no reason why "volunteer observers" couldn't do this from the comfort of their own homes. That kind of multiplication of human resources makes a lot of sense.

I will be building and improving drones this year. It will take time to get a good drone for SAR, but not impossible. Im sure there other aircraft that are also suitable for the task.

It is true that drones pose a risk to light aviation. They need to be flown low (lower than 400 feet above ground level). There are new restrictions on their use coming from the FAA in the next year or two. It really boils down to "professional" use of the drones. Like any other tool, you can use them well or abuse them.

Pete

Not sure what your engineering and flight systems background is for building drones so you might want to check out either the Walkera QR X350 or the DJI Phantom quadcopters. These consumer level "drones" are actually quite functional for the price. In fact, I have been thinking for a while to get the Phantom for our GoPro2 camera as having flight video of some of our adventures is appealing.

There a plenty of videos on YouTube of both quacopters that demonstrate their functionality. Also there a lot people who mod these for even better functionality which can be found with a Google search.

Yes I will be one more person developing drones. :-)
I have uses for the drones, so it's a definite area of activity for me.

The DJI copters are pretty nice. Flight time is better than average, and flight controls are supposedly good ( but I have not flown one myself). The main drawbacks are that the signal strength limits the DJI unit to about 300 feet range. Also, some users have had "fly always" with the DJI's. Meaning the unit went up into the sky, and didn't automatically fly back when commanded to do so. So your $1000 drone is gone ... Just gone. Not all these problems were operator error, so there are some bugs in the system that they are still working out. Just the same the DJI's are pretty nice. :-)

Good luck,
Pete

Very nice -- the DJI Phantom w/ GoPro would seem to be a good contender for an airborne scout. Would the "Phantom 2 Vision with 14MP Video Camera" (amazon.com -- see comparison chart down the page ) be a better option for SAR? Streaming WiFi while airborne would have its advantages over record/playback. A bit more money but that would be a great capability. YMMV

I like that it returns to start point if control signal is lost.

I have been researching these quadcopters for quite awhile. The range is about 300 meters (984 feet)...not 300 feet.

In the case of the original Phantom, many of the fly away problems were caused by 2.4GHz wifi signals in urban settings interfering with the Phantom flight controls. For my use, this would be a non-issue as I have no plans to buzz around the neighborhood airspace. On the other hand, a majority of fly away problems can be attributed to user error. This very much echoes the same real world of aviation; no matter how much technology any flight aircraft has, it cannot always account for, nor fix user error.

I researched this model also. I will have to see if and when I purchase the original Phantom if my flying skills are good enough. I can barely handle wrecking a $500.00 quadcopter but $1100.00 is a different story...

first thing to test would be the GPS return to start point mode. See how well it lands itself. With that all you need to do is short training flights to do take-offs, simple maneuvers and camera pointing (which is why I like the real-time video streaming).. When it's time to come back just kill the link and it should come back on its own

Thanks for the clarification on range for the DJI ... 300 yards is a lot better than 300 feet.

Another drawback is that the copters are really only useful in calm stable air ... Definitely not storms, strong winds, and night conditions (needs low light camera). So that weather limitation is significant. But the DJI looks to be very portable and could be launched from whatever position a SAR team member was located. This could really make searches of off- trail locations an easier proposition.

I like the idea of overflights with an HD camera a lot. As in ... Miles of video footage. That requires a different type of drone. But the videos could be sent to observers in their homes for scanning. So this type of crowdsourcing has excellent potential.

Pete

You wanna outsource any of your work? Maybe we can work something out that is, um, mutually beneficial???

My primary interest in drones is not SAR drones. But this thread did lead me to think about the value of doing a better SAR drone. The main drawback that I see to such work is that the SAR drone needs to be up in the air for a long time ... A long loiter time over the rescue area. Hence it should use an engine running on combustible fuel. That is OK from a tech standpoint. But any drone can crash. So the real hangup is that if an SAR drone goes down, it should not start a fire. If the drone crashes and causes a forest fire, that is a counterproductive outcome ... To say the least. This concern is not a major one, but it can't be ignored.

On the positive side ... The drone could be equipped with various sensors - making it pretty useful to the SAR team.

I really do like the idea of collecting a lot of HD video footage and then passing it to an "observer team" to scan very carefully. That really would add extra eyes and brains to the search team, without needing to put extra people on foot on the ground.

We should keep thinking about this.

Pete

Interesting discussion.

I don't pretend to be a SAR guy, but general crowdsourcing strikes me as highly dubious. It would take a high degree of on-the-ground familiarity with the type of terrain, plus typical survival behaviours, to interpret aerial photos or video footage in an effective way. The observer needs to have a practised intuition for filtering out the normal and homing in on what looks out of place.

For example, I would be useless looking at swamp, jungle, desert, and coastal rainforest; but better with aspen parkland, boreal forest, or the Northern Rockies.

There are also a lot of terrains where seasonal snowfall, extremely spiky terrain, or the sheer density of vegetation changes everything. Shelters of natural materials are perfectly camouflaged in some situations. An on-the-ground searcher could almost trip over them before they can be seen.

It may be that, in time, machine algorithms will be more effective in spotting anomalies than human observers. This is comforting, I suppose, as the machines love us for giving them something to do (other than plot world domination).

My 2c.

If someone could develop a chemical sensor as sensitive as a canine's that could be put on a drone along with the video. Find a smell that's "familiar" and follow it. Better if you can do that at a height with better lne-of-sight and able to skim above the brush and then immediately twist 180 degrees when the smell is lost. just a wild out-of-the-box thought. It's early, time for coffee -- there's a great smell, I wonder how far away a dog can smell coffee...

Be sure and equip that drone with a small keg of brandy...

Hikermor ...
It's interesting that you say that.

I was just thinking overnight about the following question. I will post it as a separate thread. If a small rescue drone could drop a light package to a victim ( package maybe weighting 2 pounds) , what should be in the package? There are probably a variety of answers to that question, and no perfect response. The immediate thing that occurred to me is that the package should contain an emergency transponder so the victim can be located again. Other than that ... I am open to suggestions. Two pound is not much wright, so the choices are hard.

Also, Doug's response is probably true. Crowdsourcing to random people probably isn't very effective. But if crowdsourcing was done to people living in the same region and terrain as the victim, that might produce better results.

Pete

The load will vary with conditions. In AZ in the summer, you will want water - lots of water. two pounds is only about a quart, which will not be enough, although way better than nothing. Some means of two way comms would be useful, so that return trips could supply pertinent supplies. Very challenging topic.

Don't expect your victim to be signaling, or really doing much of anything,at least if my experience is representative. Probably the victim is relatively or completely inexperienced, likely either dehydrated (maybe even thirsty), confused, disoriented, or injured/impaired in many possible ways. They are likely to be inert, and very likely lying beneath whatever cover the area affords....

I find this thread interesting, because, frankly, it could be pertinent to me. I am definitely not as young as I used to be, and my range is increasingly limited. But I could do a reasonably competent job of watching a screen, especially in areas I know fairly well. There are lots of geezers like me out there who would form a fairly effective resource.

No need for the world to sit around waiting for me. The Army has developed this cool little RQ-11 drone that looks quite good for searching remote areas. It is hand launched ... You just start it and throw it by hand. Brilliant!

http://www.avinc.com/uas/small_uas/raven/

The only drawback is that I don't know what these things cost. But SAR teams could get one of these pretty quickly if their budgets allowed for it. As an added tool it would be very useful.

Hikermor: very good point about including two-way comms in a package dropped to the ground. Also, it's a good thought that the victim could be in bad shape and might not be able to do very much.

One possible solution to this is to use a two-drone rescue system. The initial drone is just a small aircraft with a camera. The follow-up drone is launched after the victim is found. The second drone is a heavier "cargo drone" that drops a 10-15 pound package. This is efficient because the cargo drone does not have to fly a long search route ... The target of the drop is already located.

Pete

SAR is like real estate - the three most important things are location, location, and location. Once you have found the victim with a UAV, you most likely will want to get an actual human, preferably trained, on scene. Your victim will probably need first aid, or at least a fairly thorough assessment. You could send in some supplies, but a priority will be boots at the scene, and the development of more information.

SAR will certainly be profoundly influenced by these developments.

Hikermor has a key point. In almost every case, if you located the subject of a search, the very next move would be to get a trained rescuer on scene ASAP. In most cases it should be possible to get a helo to any location located by a drone (though that might involve winching someone down from a hover). I would think it would be an very uncommon situation where one might send in supplies before a rescuer. Possible, but unusual.

This brings up another point. The above discussion seems to assume that the subject has been positively located. A very likely case is that observation by drone locates a possible subject, or perhaps only a possible clue. A bit of color showing through the vegetation, is it the subject, a bit of clothing the subject discarded, or just some litter? A shape in the shadows that looks vaguely human? A few degraded footprints in a muddy patch, are they human or just a critter? I think almost any case that one spotted something interesting from a drone, the IMT would want to get some highly trained boots on the ground ASAP.

Are any of these drones using IR/thermal sensors or just visual spectrum? Once you go to the trouble of putting up a sensor package, make it broad spectrum -- IR & visual. I'd even examine the benefit of it being able to ping a cell phone. Even if it can't actually locate a cell phone, a ping in the wilderness is a datum.

Russ. ... Good suggestions. I'm sure that the military uses a variety of sensors, including thermal. Likewise, some sheriffs departments with a lot of resources have good thermal sensors mounted on helicopters.

The challenge for SAR drones will be to get these vehicles reliable enough that expensive kit can be mounted on them. Otherwise if the drone goes down, the loss in hardware is expensive.

Another challenge is that every additional sensor adds extra weight to the vehicle. And each sensor also requires power, which adds batteries and even moe weight. It's a feasible problem for UAV airplanes, but a tough problem for small copters where the weight carrying capability is not high.

More work needed. :-)

ASKAR ... Yes very good point. Drones might give clues. They wouldn't necessarily give a guaranteed result - unless they spotted someone waving, or clearly marked distress signals on the ground. On the positive side, if a drone operator thinks he/she has found something, a good approach is to make more overflights at lower altitudes and inspect closely from various angles. Also, the drone could drop a flare or blast a loud horn ... Something to get the attention of a victim on the ground.

Pete

The thing is, a well administered operation wouldn't use drones instead of human teams,or dogs, or any other technique. In the best case, all of these modes will coordinate and work together. The drone will see something interesting and a team will follow up, etc.

BTW, I can recall only one instance in which the victim was attempting any kind of signal.



the group was using a signal mirror and it did result in a team responding to their location.These days,most folks are going to use their cell phones, if they can. If a drone UAV is responding to a cell phone call,it might be possible to load it up with supplies.

Well ... This has been an interesting discussion. One of the better threads on ETS. It's certainly given me plenty of food for thought. I hope some constructive things develop out of it. It would be nice to see some of the ideas here put into actual practice in the real world.

Pete

I agree with you completely. Looking at your link, there are a whole bunch of interesting gadgets that one could integrate into SAR, among other things. A lot of discussion about peaceful application of drones has centered around privacy issues, a very legitimate question. SAR would certainly be a most positive use of this technology, and the sooner,the better.

Please note that Alaska beat out Ventura County, California, as a location to develop civilian applications. The weather is fine down here, now...What's it like up in AL?