Future ASW

[donald_of_tokyo]

...But officially speaking,
The unit cost is £15m,
And we could sign a contract with GA for +25 or whatever for ~£400m

But having 25 airframes, which cannot fly (control system needed), which carry no sensor, no armament, neither the software to handle them, with no technical support is for what? Of course, because of patent, we can not "modify" it by our own?

[new guy]

...But officially speaking,
The unit cost is £15m,
And we could sign a contract with GA for +25 or whatever for ~£400m

But having 25 airframes, which cannot fly (control system needed), which carry no sensor, no armament, neither the software to handle them, with no technical support is for what? Of course, because of patent, we can not "modify" it by our own?

One could say the same about F-35 or typhoon, can't operate it without an airbase. Or even a car, no road, no gas / electricity, can't use it. But people still value unit cost, be that £100m or £50k.

[donald_of_tokyo]

One could say the same about F-35 or typhoon, can't operate it without an airbase. Or even a car, no road, no gas / electricity, can't use it. But people still value unit cost, be that £100m or £50k.

Very good point, I agree. So, unit cost (actually, in many cases, many people uses "average cost" in place of unit cost) must be carefully handled.

On escorts, initial purchase cost covers a large fraction. Actually, its life cost will be about a double of the purchase cost. Also, what kind of support is included in the "program cost" differs nation to nation, and even project to project. But still, comparing them will give us a good indication of their inherent cost.

On UAV, it is clear that the airframe cost is a minor fraction. I understand that the "airframe cost of MQ-9B" is similar to the "airframe cost of E-7A and/or P-8A". B373 airframe is not that expensive and the E-7A and/or P-8A "unit cost" is dominated by its sensors and analysis systems (and its systems integration). So, I understand saying "£15m airframe cost of MQ-9B is cheap" is nearly equivalent to "B737 airframe cost of E-7".

[mrclark303]

One could say the same about F-35 or typhoon, can't operate it without an airbase. Or even a car, no road, no gas / electricity, can't use it. But people still value unit cost, be that £100m or £50k.

Very good point, I agree. So, unit cost (actually, in many cases, many people uses "average cost" in place of unit cost) must be carefully handled.

On escorts, initial purchase cost covers a large fraction. Actually, its life cost will be about a double of the purchase cost. Also, what kind of support is included in the "program cost" differs nation to nation, and even project to project. But still, comparing them will give us a good indication of their inherent cost.

On UAV, it is clear that the airframe cost is a minor fraction. I understand that the "airframe cost of MQ-9B" is similar to the "airframe cost of E-7A and/or P-8A". B373 airframe is not that expensive and the E-7A and/or P-8A "unit cost" is dominated by its sensors and analysis systems (and its systems integration). So, I understand saying "£15m airframe cost of MQ-9B is cheap" is nearly equivalent to "B737 airframe cost of E-7".

I have to say as a holistic approach, we first need 3 additional Poseidon and possibly a reinstatement of the missing two E7's.

Having fixed the basics, we should absolutely expand our Protector fleet, purchase another 20 odd systems, with the necessary ground control expansion and have the ability to reconfigure the drone as needed, between baseline Protector, to Sea Guardian and ( if possible) with the wing kit to Mojave Carrier deployable capability.

Have a pool of entirely reconfigurable airframes, we can alter as operational profiles dictate, from RAF long endurance patrol and precision strike, to land based / Carrier deployed ASW and possibly carrier organic AEW??

I would add, to make them and any other large UAV fully ( and safety ) operational on a QE class, a few mods would be needed in my opinion

First, a simple low energy arrestor system ( plus hook for the UAV obviously) and a reconfiguration of the flight deck, bringing in an angled runway, with the necessary extension to port and re ballast.

The moment you reintroduce ' conventional landing ' to a Carrier, even apparently low speed, you bring back the safety element needed by the angled runway.

We invented the angled deck after all and you don't want a heavy out of control UAV running amok straight down the middle of your flight deck if it malfunctions!!

[SW1]

The tail hook addition to a UAV with a large centre line radar pod and pusher prop will be entertaining to see.

UAVs of this size and information gathering capability suck in a lot of manpower. I think a lot of manpower supporting reaper was contractor based and for protector it will be service personnel based hence costs associated with training infrastructure and support services being stood up in the U.K.

[shark bait]

Yep, even take off and landing was handled by contractor's on the ground. After take off control was handed to RAF Waddington via satellite, and then handed back to a local team for landing.

Lots of people are needed to operate complex drones like Reaper. On a carrier it'll be almost as difficult as operating F35.

Recently GA have been working to increase automation to streamline operations, which I expect is part of the RAFs protector.

[Poiuytrewq]

ASW 2040 will be a systems of systems and looking at each facet of the system in isolation perhaps underestimates the true potential of the unmanned technological revolution.

Multiple ASW capable MALE STOL drones operating with a group of armed XLUUV’s could be a real handful for a SSN in the decades to come.

[Tempest414]

ASW 2040 will be a systems of systems and looking at each facet of the system in isolation perhaps underestimates the true potential of the unmanned technological revolution.

Multiple ASW capable MALE STOL drones operating with a group of armed XLUUV’s could be a real handful for a SSN in the decades to come.

So how will AI help with say a type 31 pulling a tail will it be able to pick up the sound of a SSN or SSK out of the clutter and highlight it to a operator

Do we think in time that a type 31 using AI could become close to say a Type 23

[shark bait]

Multistatic sonar is more important for the above application, and AI could help process the huge amount of data multistatic systems collect.

In a multistatic system a T31 could pull a powerful active sonar, essentially illuminating the underwater environment to be picked up by other passive sensors in the network.

AI is not going to make the T31 capable like a T23, but new technologies allow the Navy to approach the problem from different angles.

[Tempest414]

Multistatic sonar is more important for the above application, and AI could help process the huge amount of data multistatic systems collect.

In a multistatic system a T31 could pull a powerful active sonar, essentially illuminating the underwater environment to be picked up by other passive sensors in the network.

AI is not going to make the T31 capable like a T23, but new technologies allow the Navy to approach the problem from different angles.

Ok I am really don't know that much when it comes to ASW or AI but to be clear if a type 31 was fitted with the same sonar fit as a type 23 we could not use AI to filter out the knowern noise of the ship there for giving a better picture

once again I am just trying to get an idea of what might be possible

[SW1]

Multistatic sonar is more important for the above application, and AI could help process the huge amount of data multistatic systems collect.

In a multistatic system a T31 could pull a powerful active sonar, essentially illuminating the underwater environment to be picked up by other passive sensors in the network.

AI is not going to make the T31 capable like a T23, but new technologies allow the Navy to approach the problem from different angles.

Ok I am really don't know that much when it comes to ASW or AI but to be clear if a type 31 was fitted with the same sonar fit as a type 23 we could not use AI to filter out the knowern noise of the ship there for giving a better picture

once again I am just trying to get an idea of what might be possible

How do u know what the noise is like of both ships. At certain speeds the type 31 maybe just as effective. There is plenty of sub hunting ships with similar arrangements doing the rounds or planned I don’t think it’s anywhere close to as black and white as people make out.

[Tempest414]

So in what areas of ASW could AI make a difference

[Poiuytrewq]

Multistatic sonar is more important for the above application, and AI could help process the huge amount of data multistatic systems collect.

In a multistatic system a T31 could pull a powerful active sonar, essentially illuminating the underwater environment to be picked up by other passive sensors in the network.

AI is not going to make the T31 capable like a T23, but new technologies allow the Navy to approach the problem from different angles.

Ok I am really don't know that much when it comes to ASW or AI but to be clear if a type 31 was fitted with the same sonar fit as a type 23 we could not use AI to filter out the knowern noise of the ship there for giving a better picture

once again I am just trying to get an idea of what might be possible

How do u know what the noise is like of both ships. At certain speeds the type 31 maybe just as effective. There is plenty of sub hunting ships with similar arrangements doing the rounds or planned I don’t think it’s anywhere close to as black and white as people make out.

More than that, why not a RB2 or even even an unmanned Damen FCS 4008 like Patrick Blackett? As the sensors get more compact the displacement of the vessels required to operate them reduces.

The deployment of an ASW system of systems validates to a certain extent the supposed capabilities of the T32.

However the T31 and Enforcer 14428 combo still looks superior as it’s difficult to see how one singleton vessel could achieve so much without creating gaps in its own defence elsewhere.

IMO the most pressing concern is not a T32 style vessel, it’s an armed XLUUV to augment SSN numbers. A XLUUV that can travel huge distances at speed, lay dormant on the seabed for weeks or months or stealthily track other underwater contacts and ultimately neutralise them if required. A XLUUV capable of operating alone, as a group or with the SSN/SSBN’s. This really should be an extension of AUKUS and would provide additional mass even before the next-gen SSN’s commission.

The ultimate result should be a next-gen ASW system of systems which is securely networked to give complete situational awareness of the sub surface domain.

[SW1]

Multistatic sonar is more important for the above application, and AI could help process the huge amount of data multistatic systems collect.

In a multistatic system a T31 could pull a powerful active sonar, essentially illuminating the underwater environment to be picked up by other passive sensors in the network.

AI is not going to make the T31 capable like a T23, but new technologies allow the Navy to approach the problem from different angles.

Ok I am really don't know that much when it comes to ASW or AI but to be clear if a type 31 was fitted with the same sonar fit as a type 23 we could not use AI to filter out the knowern noise of the ship there for giving a better picture

once again I am just trying to get an idea of what might be possible

How do u know what the noise is like of both ships. At certain speeds the type 31 maybe just as effective. There is plenty of sub hunting ships with similar arrangements doing the rounds or planned I don’t think it’s anywhere close to as black and white as people make out.

More than that, why not a RB2 or even even an unmanned Damen FCS 4008 like Patrick Blackett? As the sensors get more compact the displacement of the vessels required to operate them reduces.

The deployment of an ASW system of systems validates to a certain extent the supposed capabilities of the T32.

However the T31 and Enforcer 14428 combo still looks superior as it’s difficult to see how one singleton vessel could achieve so much without creating gaps in its own defence elsewhere.

IMO the most pressing concern is not a T32 style vessel, it’s an armed XLUUV to augment SSN numbers. A XLUUV that can travel huge distances at speed, lay dormant on the seabed for weeks or months or stealthily track other underwater contacts and ultimately neutralise them if required. A XLUUV capable of operating alone, as a group or with the SSN/SSBN’s. This really should be an extension of AUKUS and would provide additional mass even before the next-gen SSN’s commission.

The ultimate result should be a next-gen ASW system of systems which is securely networked to give complete situational awareness of the sub surface domain.

Why would a xluuv need to travel huge distances at speed? Unless you mean flown by air transport. In my mind a xluuv needs to operate in the deployment sense like the submarine rescue sub. Can be deployed from any ship with a dock or a frame and predominantly then operates at choke point locations.

[SW1]

So in what areas of ASW could AI make a difference

In the area on analysis of data. It then flags up specific point s of interest to be investigated by people

[Caribbean]

Where could AI make a difference to ASW?

I think the answer to that lies in the fact that we aren't actually seeing anything that exhibits genuine AI yet. What we see today really falls under the mantle of Machine Learning, which itself is just advanced pattern recognition, using massive libraries of data.

As I understand it, ASW lies in an operators ability to recognise certain patterns, after adjusting/ allowing for a very large number of variables (heading, velocity, salinity, depth, temperature etc) as representing a contact. Current "AI" is based on doing exactly that sort of task, winnowing out all the stuff that is irrelevant to the task and identifying and ranking those combinations that match up to a specific course of action.

Current computer systems do that to a certain extent, but their coding depends on the programmer already knowing what patterns to look for. The power of AI lies in it's ability to identify patterns that have not yet been recognised as valid.

I'm explaining it poorly, but I hope you get the idea

[shark bait]

if a type 31 was fitted with the same sonar fit as a type 23 we could not use AI to filter out the knowern noise of the ship there for giving a better picture

This can be done right now without AI. If the Navy strapped a sonar to the T31 it would work fine in active mode where the signal to noise ratio is much higher, and the frequency is know making filtering and processing easier.

However it's never going to work well in passive mode with or without AI.

[shark bait]

A XLUUV that can travel huge distances at speed

What are you classing as huge distances?

In general, submarines can travel huge distances slowly, or short distances fast. Only SSNs can do both at the same time, which is out of scope for XLUUV

[Tempest414]

if a type 31 was fitted with the same sonar fit as a type 23 we could not use AI to filter out the knowern noise of the ship there for giving a better picture

This can be done right now without AI. If the Navy strapped a sonar to the T31 it would work fine in active mode where the signal to noise ratio is much higher, and the frequency is know making filtering and processing easier.

However it's never going to work well in passive mode with or without AI.

Please can you go into why you think this I am just trying to get my head around this

[Poiuytrewq]

A XLUUV that can travel huge distances at speed

What are you classing as huge distances?

In general, submarines can travel huge distances slowly, or short distances fast. Only SSNs can do both at the same time, which is out of scope for XLUUV

My point is what can we expect XLUUV’s to be able to achieve by 2040? Much more than today.

IMO the XLUUV’s will split into two distinct groups.

The first will be around 15m to 25m operated directly from suitable vessels such as the Amphibs and multi role vessels. Transit speeds and endurance will be provided by the host vessel.

The second will be much larger, perhaps up to 50m with a surfaced displacement of around 1500t. These XLUUV will be independent of other vessels with a mix of lithium batteries and AIP propulsion. The range could be up to 6000nm to 8000nm with a top speed submerged of 20kts and an endurance of around 6 months. Similar to SSK performance but with increased endurance. The armament could range from torpedos to VLS cells. All for a fraction of the cost of a SSN.

The UK needs to get to the forefront of this revolution asap.

[SW1]

A XLUUV that can travel huge distances at speed

What are you classing as huge distances?

In general, submarines can travel huge distances slowly, or short distances fast. Only SSNs can do both at the same time, which is out of scope for XLUUV

My point is what can we expect XLUUV’s to be able to achieve by 2040? Much more than today.

IMO the XLUUV’s will split into two distinct groups.

The first will be around 15m to 25m operated directly from suitable vessels such as the Amphibs and multi role vessels. Transit speeds and endurance will be provided by the host vessel.

The second will be much larger, perhaps up to 50m with a surfaced displacement of around 1500t. These XLUUV will be independent of other vessels with a mix of lithium batteries and AIP propulsion. The range could be up to 6000nm to 8000nm with a top speed submerged of 20kts and an endurance of around 6 months. Similar to SSK performance but with increased endurance. The armament could range from torpedos to VLS cells. All for a fraction of the cost of a SSN.

The UK needs to get to the forefront of this revolution asap.

By 2040 I would think we would hope to have something that words.

What you ask for in your second request hasn’t been achieved on land, on the sea or in the air so I’m think it’s wildy optimistic that it will happened underwater where challenges particularly in communication are more difficult.

[Poiuytrewq]

What you ask for in your second request hasn’t been achieved on land, on the sea or in the air so I’m think it’s wildy optimistic that it will happened underwater where challenges particularly in communication are more difficult.

Not really. The practical technology exists now.

The next evolution needs to be the communication element. That’s the crucial technological advance that will take time to progress satisfactorily.

[SW1]

What you ask for in your second request hasn’t been achieved on land, on the sea or in the air so I’m think it’s wildy optimistic that it will happened underwater where challenges particularly in communication are more difficult.

Not really. The practical technology exists now.

The next evolution needs to be the communication element. That’s the crucial technological advance that will take time to progress satisfactorily.

What technology has allowed a 1500 t submarine to operate autonomously for 6 months? Let alone at a range of 6000nm.

Come to that has any ship/submarine operated autonomously for a weeks let alone 6 months

[Poiuytrewq]

What technology has allowed a 1500 t submarine to operate autonomously for 6 months? Let alone at a range of 6000nm.

Come to that has any ship/submarine operated autonomously for a weeks let alone 6 months

The Boeing Echo Voyager already has a range of 6500nm and and an endurance of months with no input from a host vessel.

https://www.boeing.com/resources/boeing ... _sheet.pdf

https://www.boeing.com/defense/autonomo ... #/solution

I am suggesting the technology to improve this performance exists in manned SSK’s and the evolution between now and 2040 has to improve on both the performance element as well as the autonomous control element.

How were XLUUV’s looking 16 years ago?

[SW1]

What technology has allowed a 1500 t submarine to operate autonomously for 6 months? Let alone at a range of 6000nm.

Come to that has any ship/submarine operated autonomously for a weeks let alone 6 months

The Boeing Echo Voyager already has a range of 6500nm and and an endurance of months with no input from a host vessel.

https://www.boeing.com/resources/boeing ... _sheet.pdf

https://www.boeing.com/defense/autonomo ... #/solution

I am suggesting the technology to improve this performance exists in manned SSK’s and the evolution between now and 2040 has to improve on both the performance element as well as the autonomous control element.

How were XLUUV’s looking 16 years ago?

That’s a 50t concept demonstration vehicle that travels at 2.5 kts. Remains to be seen if it has proved out in real world we shall see. There is a huge jump to something at 1500 t.

There were things like Remus underwater systems for mine detection and the like back then