Man overboard - Drones with infrared cameras?

[AmazedByCruising]

It seems that quite a lot of people either fall or jump overboard. Also, it seems that even if the crew is aware, finding them is not an easy task and time is very critical.

 

Maybe I'm thinking a little bit too much out of the box here, but wouldn't drones be a perfect solution? They are very cheap, I guess $2,000 for this goal is not that much off. (Or $20.000 after all inevitable rules that the industry would have to adhere to because of the bureaucrats who apparantly think cruise lines actually like their guests to die while on holiday), they can be fitted with infrared cameras, and they can be in the air within minutes to fly back a few miles, long before the ship is able to turn around and go back.

 

I'd imagine it like this: there are about 10 drones that can be put on some upper deck as soon as the alarm sounds. Also, about 50 crewmembers are trained to operate them so there are always enough people awake to do so. As soon as the man overboard alarm is given they are airborne within 5 minutes to scan the area where the ship was a few minutes ago.

 

Maybe some IT might work here, automatically sending them to where the ship was just before someone entered the water and finding hot spots, where real people would decide if the findings are a drowning person or just a dolphin saying hello.

 

Alarm buttons on the outside could send a drone into the air immediately, to hold position and scan the area.

 

Heavy weather conditions might be a problem, then again none of the cases of people going overboard I read about were due to bad weather.

Edited October 30, 2014 by AmazedByCruising

[paul929207]

Am I right that you want to put 10 drones costing $20,000 each on every ship? And this is to help find people who have jumped off on purpose?

[cb at sea]

No one "falls" off accidentally, unless they're drunk as hell. Hardly an "accident"!

[Treven]

Am I right that you want to put 10 drones costing $20,000 each on every ship? And this is to help find people who have jumped off on purpose?

I agree, certainly not worth the expense.

[fortinweb]

It seems that quite a lot of people either fall or jump overboard. Also, it seems that even if the crew is aware, finding them is not an easy task and time is very critical.

 

Maybe I'm thinking a little bit too much out of the box here, but wouldn't drones be a perfect solution? They are very cheap, I guess $2,000 for this goal is not that much off. (Or $20.000 after all inevitable rules that the industry would have to adhere to because of the bureaucrats who apparantly think cruise lines actually like their guests to die while on holiday), they can be fitted with infrared cameras, and they can be in the air within minutes to fly back a few miles, long before the ship is able to turn around and go back.

 

I'd imagine it like this: there are about 10 drones that can be put on some upper deck as soon as the alarm sounds. Also, about 50 crewmembers are trained to operate them so there are always enough people awake to do so. As soon as the man overboard alarm is given they are airborne within 5 minutes to scan the area where the ship was a few minutes ago.

 

Maybe some IT might work here, automatically sending them to where the ship was just before someone entered the water and finding hot spots, where real people would decide if the findings are a drowning person or just a dolphin saying hello.

 

Alarm buttons on the outside could send a drone into the air immediately, to hold position and scan the area.

 

Heavy weather conditions might be a problem, then again none of the cases of people going overboard I read about were due to bad weather.

 

Where do you get "quite a few people"? Only a handful of these occurrences happen each year. Most are suicides. Rare is the person who simply "falls overboard" by accident. Considering how many millions of people cruise each year, it isn't a big problem. Worldwide, more people get struck by lightening than disappear off a cruise ship. Should we also force everyone who goes outside to wear a full metal suit to prevent being killed by lightening?

 

Seems as if you are a fan of drones and are looking for them to be a solution to a problem that hardly exists.

[Waynetor]

It does not matter if the reason is accident or suicide. The ship still has to stop and do a search. There are also false alarms where someone thinks someone went over.

Just having a cruise ship stop and retrace it path is costly, if the coast guard gets involved the cost goes up higher. If the search takes enough time a port of call can be missed or shortened costing the port as well as the cruise line money.

Drones could reduce the search time thus reducing other cost. Bulk purchasing could reduce their cost.

[joespin]

Where do you get "quite a few people"? Only a handful of these occurrences happen each year. Most are suicides. Rare is the person who simply "falls overboard" by accident. Considering how many millions of people cruise each year, it isn't a big problem. Worldwide, more people get struck by lightening than disappear off a cruise ship. Should we also force everyone who goes outside to wear a full metal suit to prevent being killed by lightening?

 

Seems as if you are a fan of drones and are looking for them to be a solution to a problem that hardly exists.

 

Now there's an idea, wearing a full metal suit to prevent being killed by lightening. :)

[SPacificbound]

If this was post topic #4 I can't imagine what 1-3 were. :eek:

 

OP; No one just accidentally falls off. It's either intentional or stupidity.

[Flatbush Flyer]

You folks may want to research "man overboard" policies and practices for unlimited tonnage vessels.

 

 

Sent from my iPhone using Forums

[Keith1010]

Relatively few people fall off a ship and rarely ever by accident.

 

Keith

[paul929207]

You folks may want to research "man overboard" policies and practices for unlimited tonnage vessels.

 

 

Sent from my iPhone using Forums

Since you seem to have research this, how many people go overboard from the ships of the major lines selling to the US? Carnival, Princess, HAL, NCL, RCI and Celebrity? In an average year?

[chengkp75]

Since you seem to have research this, how many people go overboard from the ships of the major lines selling to the US? Carnival, Princess, HAL, NCL, RCI and Celebrity? In an average year?

 

I don't think FF is commenting on the number of cases, but differing with the OP as to what the practices and procedures are for ships when a man overboard is reported.

 

For the OP, regardless of whether heavy weather is involved, when you are looking for something the size of a basketball (the head is the only thing that is visibly above water) in the ocean, even small waves and swell make it very difficult. Your drone would have to have GPS installed and be capable of transmitting its position as well as video back to the ship. I don't think that a casually trained person could control a drone sufficiently well to be able to see a person in the water, especially since most instances that DO occur happen at night.

 

For the poster who said the USCG would increase the cost of searching for someone, that is just not true. USCG SAR missions are undertaken at public expense. And you mentioned false alarms. How would a drone help in this situation? Would the ship just blithely sail AWAY from the suspected incident until the drone determined there wasn't someone there? (and how do you prove that negative?) Even for a false alarm, the ship will slow, turn, and steam back. And the same applies to an instance where it is known that someone went overboard, the ship wouldn't keep sailing, it will slow, turn, and return, so the cost would be the same.

 

While having multiple drones in the sky would place more eyeballs on the search area, as I've said, I don't think other than a military trained drone operator would be able to control, search, and identify a person floating without high visibility clothing or floatation equipment in anything other than a glass smooth sea.

[RSLeesburg]

I don't think FF is commenting on the number of cases, but differing with the OP as to what the practices and procedures are for ships when a man overboard is reported.

 

 

 

For the OP, regardless of whether heavy weather is involved, when you are looking for something the size of a basketball (the head is the only thing that is visibly above water) in the ocean, even small waves and swell make it very difficult. Your drone would have to have GPS installed and be capable of transmitting its position as well as video back to the ship. I don't think that a casually trained person could control a drone sufficiently well to be able to see a person in the water, especially since most instances that DO occur happen at night.

 

 

 

For the poster who said the USCG would increase the cost of searching for someone, that is just not true. USCG SAR missions are undertaken at public expense. And you mentioned false alarms. How would a drone help in this situation? Would the ship just blithely sail AWAY from the suspected incident until the drone determined there wasn't someone there? (and how do you prove that negative?) Even for a false alarm, the ship will slow, turn, and steam back. And the same applies to an instance where it is known that someone went overboard, the ship wouldn't keep sailing, it will slow, turn, and return, so the cost would be the same.

 

 

 

While having multiple drones in the sky would place more eyeballs on the search area, as I've said, I don't think other than a military trained drone operator would be able to control, search, and identify a person floating without high visibility clothing or floatation equipment in anything other than a glass smooth sea.

 

 

Chengkp - can you comment on the RFID bracelets that RCI plans to use on Quantum (and to which many on this board are so opposed to using). But I wonder if these would be helpful in quickly locating a passenger in the water.

 

 

Sent from my iPad using Cruise Critic Forums mobile app

[Aquahound]

For the poster who said the USCG would increase the cost of searching for someone, that is just not true. USCG SAR missions are undertaken at public expense.

 

I deleted a lot for your post to cut this down a little, but this and all else you said is right on the money.

 

Chengkp - can you comment on the RFID bracelets that RCI plans to use on Quantum (and to which many on this board are so opposed to using). But I wonder if these would be helpful in quickly locating a passenger in the water.

 

 

 

RFID bracelets aren't going to do a whole lot of good once they are outside of the sensors of the ship.

 

However, I do know cruise ships are currently in the process of installing new MOB detection systems that register falling heat signatures. They were installing this system on Zuiderdam during my cruise this past summer and I had the chance to talk to the men installing it. It's sophisticated enough to differentiate falling persons from falling deck chairs.

 

But that said, recovery will always be a challenge, especially at night.

[YubaSutter]

Where do you get "quite a few people"? Only a handful of these occurrences happen each year. Most are suicides. Rare is the person who simply "falls overboard" by accident. Considering how many millions of people cruise each year, it isn't a big problem. Worldwide, more people get struck by lightening than disappear off a cruise ship. Should we also force everyone who goes outside to wear a full metal suit to prevent being killed by lightening?

 

Seems as if you are a fan of drones and are looking for them to be a solution to a problem that hardly exists.

 

I don't think a full metal suit would prevent one from being struck by lightening. I seem to remember metal attracts lightening.

[fortinweb]

I don't think a full metal suit would prevent one from being struck by lightening. I seem to remember metal attracts lightening.

 

That's the point of a Faraday suit. It attracts lightning and dissipates it into the ground. Illusionist David Blaine used one for a stunt he created in 2012.

 

 

Other links to more information on this stunt:

 

 

http://www.nydailynews.com/news/national/david-blaine-begins-electrifying-stunt-article-1.1176526

 

Edited October 31, 2014 by fortinweb

[chengkp75]

I deleted a lot for your post to cut this down a little, but this and all else you said is right on the money.

 

 

 

RFID bracelets aren't going to do a whole lot of good once they are outside of the sensors of the ship.

 

However, I do know cruise ships are currently in the process of installing new MOB detection systems that register falling heat signatures. They were installing this system on Zuiderdam during my cruise this past summer and I had the chance to talk to the men installing it. It's sophisticated enough to differentiate falling persons from falling deck chairs.

 

But that said, recovery will always be a challenge, especially at night.

 

Yes, RFID is very short range, which is why they've got to nearly brush past you to clone your credit card.

 

I'm still not convinced about the man overboard sensors. Ex-Techie was pushing this on the Disney forum a while back. When one of their spokespeople came on the thread and claimed the sensor could distinguish "mass" using lasers, I questioned how that worked. He retracted to say "size" or "volume", so if their own people can't get it right (kind of like cruise line employees), I will have doubts. I wonder, and I believe I asked this on that thread, if the sensor could differentiate between a person and a burning trash can (similar size and possibly heat signature). Anyway, if the ships feel that it is worth installing this stuff (at least until it stops working from the sea environment), good for them, but as you say, I don't think it will increase the survival rate.

[AmazedByCruising]

I wasn't trying to find new ways to use drones, the fact that our ship was filmed while in harbour by a drone sparked the idea.

 

The costs of drones maybe high, but if it does work a lot of money is saved as well.

 

And I think it might work, like in this case: http://venturebeat.com/2013/05/10/mounties-save-their-man-with-a-draganflyer-uav-drone-aircraft/

[RSLeesburg]

RFID bracelets aren't going to do a whole lot of good once they are outside of the sensors of the ship.

 

However, I do know cruise ships are currently in the process of installing new MOB detection systems that register falling heat signatures. They were installing this system on Zuiderdam during my cruise this past summer and I had the chance to talk to the men installing it. It's sophisticated enough to differentiate falling persons from falling deck chairs.

 

But that said, recovery will always be a challenge, especially at night.

 

 

Yeah, I was not sure how much range those had. Although, if they are monitored continuously, I wonder if you could detect and maybe alert someone when one dropped off the system. Then security could see where it was last registering, and if there is concern about someone jumping, maybe review camera footage of that area. Then again, if you were going to jump, I would think you could just take the bracelet off before you did.

 

 

 

Sent from my iPad using Cruise Critic Forums mobile app

[chengkp75]

I wasn't trying to find new ways to use drones, the fact that our ship was filmed while in harbour by a drone sparked the idea.

 

The costs of drones maybe high, but if it does work a lot of money is saved as well.

 

And I think it might work, like in this case: http://venturebeat.com/2013/05/10/mounties-save-their-man-with-a-draganflyer-uav-drone-aircraft/

 

It's possible they may work, but note that the drone started from a known GPS position and searched a 200 meter radius before finding the man. Far different from a case where the position the person went overboard might not be known, where the current and wind can cause unexpected drifting of the person, and where the person is possibly being covered intermittently by cool sea water (and water transmit heat from the body 6 times faster than air), so the person in the water will be far closer to ambient temperature.

[AmazedByCruising]

At 24 degrees Celsius (75 Fahrenheit) you're supposed to die. https://en.wikipedia.org/wiki/Human_body_temperature#Cold

Yet, at below freezing temperatures, people were alive for more than 15 minutes on the Titanic https://en.wikipedia.org/wiki/RMS_Titanic#Sinking

Which leads me to believe that if the water is not too warm and the response is quick enough (and drones would be in the air within minutes), a person could easily be seen as a bright light using infrared cameras.

 

Here's a possible scenario I think. A guest sees another guest jump/trip/whatever overboard (tripping is not that impossible, even the captain of the Costa Concordia tripped while his ship had some bad luck. Having a bit more luck than his ship he fell right into a lifeboat). First thing the officer should ask is: how long ago? He might answer "3 minutes, I guess". Some time between 1 and 10 minutes would be realistic. Let's say the ship sails at 16 nm/30km per hour. As the ship's track is known, the area to be searched would then be between 0.5 km "ago" and 5 km, 4.5 km (2.8 land miles) in length. Accounting for drifting (wild guess): 200 meters wide. 10 drones, each scanning at a 10 meter radius should be able to find the unfortunate passenger really quick. If there are cameras on board and the reporting guest can be seen running to the front desk, the exact time of the incident could of course be calculated much more precisely, leading to a much smaller area to be searched. A "When you see someone jump, look at your watch first, then report" policy might help as well.

 

Anyway, I have been on a train that stopped for 2.5 hours because someone decided to end their life really quickly. A man overboard situation must feel similar to guests and crew, and it's not a very nice feeling. When officers "would rather not talk about the alarm last night", I don't think I'd be enjoying the rest of the holiday that much. If the alarms went off at 3:30 AM getting thousands of guests wide awake and scared, and the event would end with the captain announcing at 4 AM that the guest was rescued and the ship had an improvised celebration with champagne in the Main Dining room, formality level: pyjamas allowed, I'm sure the sales department of the company would have a good reason to open up some champagne as well.

[chengkp75]

At 24 degrees Celsius (75 Fahrenheit) you're supposed to die. https://en.wikipedia.org/wiki/Human_body_temperature#Cold

Yet, at below freezing temperatures, people were alive for more than 15 minutes on the Titanic https://en.wikipedia.org/wiki/RMS_Titanic#Sinking

Which leads me to believe that if the water is not too warm and the response is quick enough (and drones would be in the air within minutes), a person could easily be seen as a bright light using infrared cameras.

 

Here's a possible scenario I think. A guest sees another guest jump/trip/whatever overboard (tripping is not that impossible, even the captain of the Costa Concordia tripped while his ship had some bad luck. Having a bit more luck than his ship he fell right into a lifeboat). First thing the officer should ask is: how long ago? He might answer "3 minutes, I guess". Some time between 1 and 10 minutes would be realistic. Let's say the ship sails at 16 nm/30km per hour. As the ship's track is known, the area to be searched would then be between 0.5 km "ago" and 5 km, 4.5 km (2.8 land miles) in length. Accounting for drifting (wild guess): 200 meters wide. 10 drones, each scanning at a 10 meter radius should be able to find the unfortunate passenger really quick. If there are cameras on board and the reporting guest can be seen running to the front desk, the exact time of the incident could of course be calculated much more precisely, leading to a much smaller area to be searched. A "When you see someone jump, look at your watch first, then report" policy might help as well.

 

Anyway, I have been on a train that stopped for 2.5 hours because someone decided to end their life really quickly. A man overboard situation must feel similar to guests and crew, and it's not a very nice feeling. When officers "would rather not talk about the alarm last night", I don't think I'd be enjoying the rest of the holiday that much. If the alarms went off at 3:30 AM getting thousands of guests wide awake and scared, and the event would end with the captain announcing at 4 AM that the guest was rescued and the ship had an improvised celebration with champagne in the Main Dining room, formality level: pyjamas allowed, I'm sure the sales department of the company would have a good reason to open up some champagne as well.

 

What you are talking about is when core body temperature reaches 75F, death is inevitable. However, temperature in the extremities will be much lower, as the body stops blood circulation to the extremities to try to maintain core temperature. Therefore, your thermal image decreases as the surface temperature of the person decreases.

 

While your scenario is perhaps feasible, I would say that most overboards are not observed (went to cruisejunkie, and in the small sample of 19 overboards in 2014, only 3 or 4 were observed), so the timeline gets much muddier, and the search area gets much bigger. Not sure how you would track the drones to ensure that you are actually searching in a complete grid, and not drifting due to crosswind, etc. The drone would have to have a radar reflector on it to be visible to the ship, and even then, there are blind windows around the ship (typically directly behind the funnel) where radar doesn't work, and marine radar typically won't differentiate a small target like a drone from 'clutter'. You would also have to have 10 trained operators, who have worked together on drone formation flying, standing by waiting to fly the drones (not likely, they will be doing another job onboard), so you will have some time while these 10 people drop what they are doing, rush to the drone control center, have others ready the drones, and then launch.

 

And unfortunately, as a fact of life in today's society, if a single report of a person going overboard is not corroborated by a second means (security camera, for instance), there will be a passenger count before the ship does much more than start to slow down.

[DirtyDawg]

It seems that quite a lot of people either fall or jump overboard. Also, it seems that even if the crew is aware, finding them is not an easy task and time is very critical.

 

Maybe I'm thinking a little bit too much out of the box here, but wouldn't drones be a perfect solution? They are very cheap, I guess $2,000 for this goal is not that much off. (Or $20.000 after all inevitable rules that the industry would have to adhere to because of the bureaucrats who apparantly think cruise lines actually like their guests to die while on holiday), they can be fitted with infrared cameras, and they can be in the air within minutes to fly back a few miles, long before the ship is able to turn around and go back.

 

I'd imagine it like this: there are about 10 drones that can be put on some upper deck as soon as the alarm sounds. Also, about 50 crewmembers are trained to operate them so there are always enough people awake to do so. As soon as the man overboard alarm is given they are airborne within 5 minutes to scan the area where the ship was a few minutes ago.

 

Maybe some IT might work here, automatically sending them to where the ship was just before someone entered the water and finding hot spots, where real people would decide if the findings are a drowning person or just a dolphin saying hello.

 

Alarm buttons on the outside could send a drone into the air immediately, to hold position and scan the area.

 

Heavy weather conditions might be a problem, then again none of the cases of people going overboard I read about were due to bad weather.

 

Interesting idea. The one thing that I have learned about technology and it's advancements over the past 50 years or so - Never say Never.;)

[AmazedByCruising]

What you are talking about is when core body temperature reaches 75F, death is inevitable. However, temperature in the extremities will be much lower, as the body stops blood circulation to the extremities to try to maintain core temperature. Therefore, your thermal image decreases as the surface temperature of the person decreases.

 

Luckily the very last extremity to be shut down is your head because nature knows that the brain is a lot more important than your left foot. So his hands might be invisible, but if the drownee (?? I'm not native to English) is still breathing air, his head should be visible.

 

 

While your scenario is perhaps feasible, I would say that most overboards are not observed (went to cruisejunkie, and in the small sample of 19 overboards in 2014, only 3 or 4 were observed)

 

OK, that would be a show stopper... Then again, how did they know about the other 15? Only after disembarkment?

 

> The drone would have to have a radar reflector on it to be visible to the ship

 

Why's that? IMHO the drone finding the passenger should just hover and maybe switch on a big light. Even my phone has a GPS function that knows where I am within, say, 10 meters, why could the drone have that?

 

>You would also have to have 10 trained operators, who have worked together on drone formation flying.

 

Yes. Yet, on the only cruise I've ever been on I've witnessed drills every single day with a lot of crew involved. Training 50 crew members to operate a drone to ensure there are 10 available 24/7 wouldn't be too hard. Actually, I would be filming such a drill and find it more interesting than the next old church.

 

> And unfortunately, as a fact of life in today's society, if a single report of a

> person going overboard is not corroborated by a second means (security

> camera, for instance), there will be a passenger count before the ship does

> much more than start to slow down

 

You mean that if a real big ship has someone jumping overboard at 2 AM, yet the front office thinks it might be a very poor joke, they actually start by knocking at each cabin to count 5000 passengers and 2000 crew members instead of turning around immediately?

[chengkp75]

Luckily the very last extremity to be shut down is your head because nature knows that the brain is a lot more important than your left foot. So his hands might be invisible, but if the drownee (?? I'm not native to English) is still breathing air, his head should be visible.Quite correct, but you are now reducing the target to sphere 6-8" in diameter.

 

 

 

 

OK, that would be a show stopper... Then again, how did they know about the other 15? Only after disembarkment? Sometimes, and sometimes a fellow traveler notices the person missing the next morning.

 

> The drone would have to have a radar reflector on it to be visible to the ship

 

Why's that? IMHO the drone finding the passenger should just hover and maybe switch on a big light. Even my phone has a GPS function that knows where I am within, say, 10 meters, why could the drone have that? Okay, but does your cell phone transmit your position to someone else? So there would need to be a GPS transmitter on the drone and the requisite receiver and display on the ship. And given your GPS accuracy of 10 meters, and your search grid of 10 meters, there could be a gap of 10-20 meters in the search pattern. You would need a military grade differential GPS.

 

>You would also have to have 10 trained operators, who have worked together on drone formation flying.

 

Yes. Yet, on the only cruise I've ever been on I've witnessed drills every single day with a lot of crew involved. Training 50 crew members to operate a drone to ensure there are 10 available 24/7 wouldn't be too hard. Actually, I would be filming such a drill and find it more interesting than the next old church.I would suggest that in order to fly a formation search, the operators would need several hundred hours training together, so that each operator knows how the others react to environmental factors, so the same crew would always need to be grouped together. Crew training in safety drills is mandatory by regulation, but is NOT paid for. The training is usually done during the crew's off hours, yet they are not entitled to additional pay for the additional hours. Getting someone to agree to take on an extra 100 hours, plus refresher training, would be difficult.

 

> And unfortunately, as a fact of life in today's society, if a single report of a

> person going overboard is not corroborated by a second means (security

> camera, for instance), there will be a passenger count before the ship does

> much more than start to slow down

 

You mean that if a real big ship has someone jumping overboard at 2 AM, yet the front office thinks it might be a very poor joke, they actually start by knocking at each cabin to count 5000 passengers and 2000 crew members instead of turning around immediately?

 

That would not be the "front office's" decision, that would be the Captain's decision, and yes, that would be, and in past cases has been the situation. Besides, with a modern cruise ship travelling at 18-21 knots, if the Captain were to turn the vessel around using one of the various types of "course reversal" turns used to return to the exact reciprocal track, the ship would experience every passenger being thrown to the floor, with injuries, and damage to virtually anything onboard that isn't bolted down. The ship needs to slow down quite a lot before it can turn around. It will generally travel 2-3 miles before it is safe to turn. The Norwegian Sky had an incident like this back in 2002 or so, when the vessel turned radically at sea speed. We disembarked 100+ passengers to hospitals in Vancouver with broken bones, and virtually every plate and glass on the ship was broken.