Exceptionally Large Autonomous Attack Submarines

In recent years there have been breakthroughs in technology that will enable to the use on very large unmanned submarines that can sink surface ships and other submarines

Such technologies are artificial intelligence. the ability of a computer to identify, track, and engage targets autonomously
Breakthroughs in Battery technology, and in fuel cell technology
Reliable and affordable satellite communications, allowing the sub to periodically come close to the surface and receive updated targeting information

It appears that large autonomous vessels will reach the size of conventional diesel electric submarines in time.
Plausible sizes for these vessels into the future would be of the order of 1000t, 2000t and even 3000t

Autonomous vessels have plusses and minuses.
One plus is the elimination of a great deal of equipment, crew berths, lighting, air conditioning, galley, life rafts, medical bay, display consoles etc, diesel engine, fuel, generators etc

The elimination of this equipment permits more space for batteries/fuel cells for propulsion. Thus mass fraction for batteries could increase from about 12 percent in a conventional diesel electric submarines, to 30 or even 50 percent in drone subs.

The use of Li Ion batteries that might have a very low, but not a zero chance of catching fire, is deemed permissible in an uncrewed submarine
Li Ion batteries have 7 times the energy density of Lead Acid batters 280 kW hours per tonne vs 40 kW hours per tonne.
The lack of crew permits long slow transit speeds to areas of operations, the low speed resulting in less power consumption
Hotel load of an unmanned sub will be lower because there is no need to cater for crew comforts
Another plus is the much lower cost by building a much simpler vessel
The vessel can be built to a lower standard and hence lower cost because human lives are not at risk

Some negatives are going to be the inability to reload torpedoes
the inability to maintain vessel systems
The reliance on software to interpret sensor information, which may not have the flexibility of the human mind
The difficulty of maintaining a battery cooling system without crew

In decades to come drones like this will become more common, what technologies are employed and what size there vessels eventually reach

Scott Elaurant

Well-Known Member
I accept the logic of your premise in that an all-(battery) electric large uncrewed submarine (SSE?) will have space saving advantages due to not needing a crew. Hence it may be possible to develop such a vessel with a range sufficient for independent action in nearby waters. For the reasons I outlined on the other thread about the potential energy density of battery storage, it will never outrange diesel submarines, with or without crew, once weapons are added.

That being said, I think this is still a very long term development project, which does not replace the need for current navies to operate SSKs and/or SSNs with support from smaller UAVs.

The development of control and communication systems for a true UAV/SSE will be a huge undertaking which will take many years. I am familiar with the time it has taken to develop self-driving systems for autonomous cars to drive from A to B, which is to say far longer (decades) than the IT people first promised. When first proposed in the late 2000s this would take "about five years". 15 years later the more optimistic CAV developers are proposing road legal (proven reliable) CAVs around 2040.

Getting from A to B on a two dimensional surface (land) is a far less complex task than a command and control system for a submarine which must navigate in 3D, detect and avoid obstacles, detect, identify and report potential adversaries, and carry out some sort of mission profile. Making decisions about when to attack against some pre-set rules of engagement would be incredibly complex. How do you repalce the human decision maker in a submarine that can sink ships undetected?
  • Thread Starter Thread Starter
  • #3
Yes all true

I agree that time frame is not imminent, it might be ten, fifteen or twenty years out.

There would need to be strict logic embedded into the software, what can you shoot, at what time,, can you shoot in certain areas and not others. Target identification will be a problem too, as it no doubt is for existing submariners, as say a warship of the same class may be used by several navies, how do you ensure you are targeting the correct vessel

CAV means car autonomous vessel?

One plus that can be factored in with high probability is the much lower unit cost, which allows multiple USVs (is unmanned submersible vessel the correct term?) to work as a team. If a target is detected passively, it may well be that one of the group uses active sonar, whilst the remaniner listen in and use the return echo for targeting information

The Australian Navy has greater difficulties than other nations in using submarines that dont have diesels or nuclear power, and that is operating at very long range.

It may well be that in this case, out of the box thinking is required.

Perhaps the UAV could tow a battery pack in a streamlined pod, for a few thousand miles, and then release the pod when close to area of operations, and once there continue with fully charged batteries

Another option might be a very simple unarmed and uncrewed tug submarine, that tows the armed vessel close to area of operations and then lets it go. Logic suggests that a large unmanned and unarmed tow vessel would be relatively quite cheap to make

It might be that some navies with less money go down the route of simpler unmanned submarines, they might have a simple and crude logic, that in wartime, shoot anything large that moves in this predefined area

some ideas to think about

John Fedup

The Bunker Group
I imagine some nations are exploring nuclear XLUUVs which would solve the range issue. They likely would be very carefully watched and deployment would be for extreme situations only (at least initially) due to the risk of nuclear material being captured somehow.

John Fedup

The Bunker Group
A new large UUV concept call MUM will be built in Germany. Its main feature is the extra width compared to other XLUUVs. It is a commercial application but will likely be watched by naval planners.



The Bunker Group
Verified Defense Pro
A new large UUV concept call MUM will be built in Germany. Its main feature is the extra width compared to other XLUUVs. It is a commercial application but will likely be watched by naval planners.
Technically it's not a commercial application.

It's a joint public-private R&D project formally conducted by Federal R&D Center Jülich with industry and non-industry partners on behalf of the government. Project Coordinator is TKMS. Total project cost for the current MUM2 phase is 35 million, of which 22 million is financed by the Federal Ministry of Economy.

Project partners for MUM with workshare are:
  • Thyssenkrupp Marine Systems : system integration, platform technology; soft- and hardware for mission planning and remote operation (through subsidiary Atlas Elektronik)
  • EvoLogics : communications swarm system (carried)
  • Fraunhofer Institute for Communication, Information Processing and Ergonomics : cybersecurity aspects
  • DLR Institute for Protection of Maritime Infrastructures : maritime traffic aspects ; legal aspects (through associate partner Federal Maritime and Hydrographical Agency)
  • University of Rostock : sub-ROVs (carried), multi-vehicle guidance control
  • Technical University of Berlin : simulation and modelling
The project is a follow-on to a previous "MUM1" phase which ran from 2017 to 2020, was also basically a TKMS project with the same partners and was similarly financed by the Ministry of the Economy to build a 1/5-scale demonstrator.
Last edited: