Works the other way as well. Uncrewed systems could turn out to be more cost effective than manned systems.
Royal Australian Navy Discussions and Updates 2.0
- Thread starter OPSSG
- Start date
U.S. Submarine Tender to Support AUKUS Pillar 1 Milestone - Naval News
Another good news AUKUS story. This relates to the arrival of USS Emory S Land in FBW. Possibly the ugliest ship in the USN, and I suspect getting towards the oldest outside of the carriers. She would never have made the news before the AUKUS deal.
She is here to conduct the very first SSN maintenace activity in FBW, making this an important milestone for other SSNs to start home basing. The article says that FSU-W (FIMA back in my day) sailors and officers are involved, supervised by a contingent from the USN.
This is good to hear as I expected this to almost entirely be contracted to ASC. Too often all the good work bypasses enlisted staff.
Another good news AUKUS story. This relates to the arrival of USS Emory S Land in FBW. Possibly the ugliest ship in the USN, and I suspect getting towards the oldest outside of the carriers. She would never have made the news before the AUKUS deal.
She is here to conduct the very first SSN maintenace activity in FBW, making this an important milestone for other SSNs to start home basing. The article says that FSU-W (FIMA back in my day) sailors and officers are involved, supervised by a contingent from the USN.
This is good to hear as I expected this to almost entirely be contracted to ASC. Too often all the good work bypasses enlisted staff.
Going Boeing
Well-Known Member
I recall an old article about the Ghost shark that indicated that most of the internal volume of the vessel is “wet” with only a small pressurised volume for the guidance controls, etc that remains dry. The wet area is where payload in the form of weapons (torpedoes, mines), sensors, etc would be carried & I suspect that they would be gravity released out of the bottom of the vessel thus they wouldn’t require complex weapon discharge systems. I would think that the design brief to Anduril Australia would be to keep them as simple as possible and thus be affordable to be produced in large numbers.
SM-3 is a kinda special case. Its not a multipurpose missile to shoot down planes, helicopters and cruise missiles.
You have a very specific threat, from a very specific high level US peer power. And you specifically want a capability to hit their high value strategic asset. Its very expensive, and built in small numbers.
For Australia, it is even more curious thing, because we aren't really facing the same threats as Korea or Japan or Poland. The kind threats we may be using it on, are perhaps even more high value and ambitious, think orbital. Or it specifically saying we are going to have an asset deployed to a region where it would be used otherwise (which with just 3 ships is pretty unlikely when the US, Korea and Japan have much more suited capability).
So for Sm-3 to go from a latent, potential thing, to a real blind extreme value asset thing is interesting. I imagine it could also be about southern hemisphere coverage. Its not an Australia specific threat, its about global capability.
With very elliptical orbits you can make space based assets very difficult to intercept in certain hemispheres. I don't know how many Aegis BMD ships are frequently deployed to the southern hemisphere, but its not many. Few would have the capability to track and coordinate such an attempt.
It would make the effort of protecting space based assets much, much harder.
I have heard that the MU90 has hard kill capability, when I google this not a lot of info comes up, which is not suprising. Some of it on the first page is me.
EuroTorp
The MU90 / IMPACT Advanced Lightweight Torpedo, is a joint development between WHITEHEAD ALENIA Sistemi Subacquei (Italy, a FINMECCANICA company),…
It would make sense why the RAN continues to persist with MU90 capability on ships. And why mk54 etc are just fine for P8 and Mh60R. There are other advantages of each type, so it may not be that clear.
I am imagining that something like the Ghost Shark would be ideal for minelaying. Fits in nicely with defence force's purchase of an undisclosed number of sea mines.
Realistically a slow moving vessel such as the Ghost Shark won't be hunting down and torpedoeing enemy ships. More likely they will be used for more mundane tasks such as minelaying. Also if plans to manufacture these mines in Australia go ahead it will give Australia a sovereign ongoing capability.
Agree fully, I am not at all against SM-3 in RAN service. I am not yet certain it’s a definite goer is all and the comment from Conroy is hardly definitive proof, despite some “curious” media reporting to that effect, based on nothing more than what was written in a transcript of a presser…SM-3 is a kinda special case. Its not a multipurpose missile to shoot down planes, helicopters and cruise missiles.
You have a very specific threat, from a very specific high level US peer power. And you specifically want a capability to hit their high value strategic asset. Its very expensive, and built in small numbers.
For Australia, it is even more curious thing, because we aren't really facing the same threats as Korea or Japan or Poland. The kind threats we may be using it on, are perhaps even more high value and ambitious, think orbital. Or it specifically saying we are going to have an asset deployed to a region where it would be used otherwise (which with just 3 ships is pretty unlikely when the US, Korea and Japan have much more suited capability).
So for Sm-3 to go from a latent, potential thing, to a real blind extreme value asset thing is interesting. I imagine it could also be about southern hemisphere coverage. Its not an Australia specific threat, its about global capability.
With very elliptical orbits you can make space based assets very difficult to intercept in certain hemispheres. I don't know how many Aegis BMD ships are frequently deployed to the southern hemisphere, but its not many. Few would have the capability to track and coordinate such an attempt.
It would make the effort of protecting space based assets much, much harder.
I have heard that the MU90 has hard kill capability, when I google this not a lot of info comes up, which is not suprising. Some of it on the first page is me.
But Hard kill isn't exactly a black and white thing. A WW1 biplane has air to air intercept capabilities, but that may not be relevant against modern threats. There is a continuum. I imagine low speed underwater drone threats may be more easily intercepted. Mu90 says navigation still possible in 3m of water. They may be able to operate as mines as well. But there is a big difference between a mk48 and a mk54. Its like a 50 cal and a 2000lb Jdam. mk48 is like ten times the warhead power and the mk54 would be more about making a small hole in thick plate than a big hole in thin plate (more like a tank round in that regard). Mk48 could probably penetrate a surface ship hull on kinetics alone, I'm not sure a mk54 would.EuroTorp
The MU90 / IMPACT Advanced Lightweight Torpedo, is a joint development between WHITEHEAD ALENIA Sistemi Subacquei (Italy, a FINMECCANICA company),…www.naval-technology.com
It would make sense why the RAN continues to persist with MU90 capability on ships. And why mk54 etc are just fine for P8 and Mh60R. There are other advantages of each type, so it may not be that clear.
StevoJH
The Bunker Group
My biggest concern with the amount of investment in such a small number of hulls, is that in a conflict if there are any attacks on Australian soil it might make it extremely hard politically for a government to deploy these assets away from our major population centres and strategic centres.
Especially with our lack of short-medium range GBAD, let alone theatre level systems such as THAAD, Patriot, SAMP/T or Aegis Ashore. Add on the very isolated nature of the more forward RAAF and RAN bases, let alone FBE, FBW, Amberley and Williamstown.
The addition of large relatively inexpensive UUV's with long endurance, which could possibly be pre-deployed into an area of operations prior to a conflict also significantly increases the risk of strikes on strategic targets that previously would have been considered to be relatively safe.
It would be a commitment in itself. And perhaps premature until the Aegis update is complete. I think your skepticism is well founded.
This may be one of those times the sort of softly put it out there and see what reaction is. Sm-3 BMD is a whole different game, and is usually much more than just about defending against a conventional war head from a missile which people paint it as.
So fits into that careful air of superpower deterrence, much like nuclear weapons themselves.
Firing SM-3 en masse into satellites is a terrible outcome for humanity. You are talking about making manned space exploration (and perhaps even regular space sats entirely) impossible after that point, perhaps for decades or centuries, perhaps forever. Particularly in highly elliptical orbits, not just low flying decaying orbits like in test fires. So it has the sort of strategic political fall out nuclear weapons have.
Firing them off, without due care and calculation, risks at blinding your own space based assets as well. I am sure the Americans would be most displeased if someone fired SM-3 off and killed all the astronauts in space, and blinded/damaged all the LEO sats both friendly and unfriendly. And made space based launches impossible for half a century.
So yeh, no cowboy stuff. Its not just a bigger SM-6. This is space cruiser stuff. Australia does a lot of space tracking, we aren't idiots in this sort of game. But it is a high stakes space, or could be.
Plenty of countries have made noises about deploying it, but basically it more as I understand like a dual key arrangement with the US, which, everyone is happy about. Japan, Korea and Poland fit into that, because deterrence. Australia doesn't.
But we do fit into knocking things out of the southern skies as part of a wider multinational effort.
Scott Elaurant
Well-Known Member
Yes that is generally the case in other forms of transport like civilian trains and aircraft and now military UAV drones. The technology for self-guidance is both proliferating and getting cheaper. Of course a warship captain has to make a lot more decisions than a cargo ship captain, but if those could be done remotely the vessel control functions of a warship could be automated. Even very large container ships now are down to crews as small as 15 to 20 people with currently available automation.
So in summary I think the question of effective control on an automated warship will determine feasibility of this concept, not cost. If you consider the level of complexity of car self-driving concepts already on the road, in cars costing under $100K, the cost would be trivial in comparison to any ship larger than a patrol boat.
Last edited:
InterestedParty
Active Member
How on earth do you code the rules of engagement for an automated weapon of war. Machines are not renowned for their common-sense judgement or humanity.
I seem to remember that there were instances during the Iraq war where pilots refused to drop there weapons because of what their eyes, ears and brains were telling them.
StevoJH
The Bunker Group
How do you code the rules of engagement for a sea mine?
Set your area of operations, set up target profiles, make sure nothing friendly enters the area unless the AUV has a way of distinguishing between friend and foe.
Unload a few of these at night from passing container ships off a major port prior to the start of a major conflict and you could cause all sorts of havoc. And unlike a mine, these move.
iambuzzard
Active Member
All we don't need is a new Skynet with fleets of robotic weapons of war getting out of control. There must be humans with ultimate control. The safeguards have to be rock solid otherwise a potential friendly fire incident would be catastrophic.
Doesn't AEGIS already have a no human in the loop fully automatic engagement mode? As well as Phalanx and active protection systems for vehicles?
If thats correct, what’s the difference?
Todjaeger
Potstirrer
Area and breadth of capability/response, as well as scale.
Take Phalanx in autonomous mode for instance. It has it's own radar and FCS which can effectively scan a comparatively narrow (i.e. not 360 degree circular horizontal or vertical planes) area, likely only for a few km's and then engage. The effective engagement range of a Mk 15 Phalanx is only about one to 1.5 km, and whilst some of the projectiles likely could travel farther, and still be potentially damaging or lethal, it would still be considerably short ranged (and thus smaller total potential area of impact) than a naval gun, never mind when compared to rocket artillery or standoff ordnance ranges.
An old programming adage, which I first heard from my father (who spent 41 years in data processing) is as follows:
Now consider the implications of, and potential impact from, an autonomous/AI-controlled weapons system with a hardware controller defect and/or a programming flaw, when this autonomous/AI system is controlling a weapons system capable of striking a large area, or responsible for securing a broad swath of ground or airspace.
A programming flaw could prompt the AI to engage 'targets' which are not legitimate, potentially leading to an ineffective use of ordnance, OTOH such a flaw could also trigger engagement of civilian targets and population centres or even blue-on-blue fraticides. We have already seen such fraticide when aircraft displayed the incorrect IFF data for the airspace they were operating in. Imagine the potential impact of an AI which reads the IFF system correctly but then determines/decides it is wrong.
AFAIK we are still a long way away from having computers and computer-controlled systems which are both robust enough and reliable enough for long-term continuous operation without human intervention. More to the point, how many times have you as a user ended up rebooting your computer because of an issue?
The version I remember from 30 years in IT.
Yes, that’s is all something to be wary and concerned with. But the breadth and scale is AEGIS though? Isn't that active worldwide with a fully automated no human in the loop mode?
If I am correct, that would mean there is many, active fully automated long range systems already, controlling multiple types of SAM, Tomahawks, VLASROC, Harppon, NSN maybe others I’ve missed?
Intriguing discussion on how much autonomy people would be willing to give to a war machine. Do we let them fight on our behalf, and how much of the final fire decision does the machine get to make.
I would view that if faced with a choice to put people or machines in harms way, we will select machines (nobody likes the imagery of dead soldiers). Machines can detect and react faster than us, so a machine that can make its own decision, is going to be superior to a machine that needs to wait for a humun authority. If you want our machines to be in the fight, let alone have the chance to win, ultimately the machines will need autonomy.
I also question the benefit of the human in the fire chain. We are a long way from having perfect decision making capabilities. Either because of error, or deliberate malevolence. We worry about a robot mistaking the wrong target, I doubt they could be worse than humans at this. We worry about robots killing civilians. Ukraine shows we are very good at that ourselves.
Australia recently purchased some really nice smart sea mines from RWM Italia. These things sit on the ocean floor listening for the right target and then activate. All by themselves. We already have the first batch. What is the difference between that and a torpedo carrying ghost shark patroling a region by itself, listening for a defined target and then firing. Our choice in this process is to release the machine or not, and the parameters we give it to define an enemy target.
Perhaps we need to invest in IFF system improvements as our mitigation.
Armed autonomous machines are going to be available this decade in numbers. Our enemies will certainly deploy them.
I would view that if faced with a choice to put people or machines in harms way, we will select machines (nobody likes the imagery of dead soldiers). Machines can detect and react faster than us, so a machine that can make its own decision, is going to be superior to a machine that needs to wait for a humun authority. If you want our machines to be in the fight, let alone have the chance to win, ultimately the machines will need autonomy.
I also question the benefit of the human in the fire chain. We are a long way from having perfect decision making capabilities. Either because of error, or deliberate malevolence. We worry about a robot mistaking the wrong target, I doubt they could be worse than humans at this. We worry about robots killing civilians. Ukraine shows we are very good at that ourselves.
Australia recently purchased some really nice smart sea mines from RWM Italia. These things sit on the ocean floor listening for the right target and then activate. All by themselves. We already have the first batch. What is the difference between that and a torpedo carrying ghost shark patroling a region by itself, listening for a defined target and then firing. Our choice in this process is to release the machine or not, and the parameters we give it to define an enemy target.
Perhaps we need to invest in IFF system improvements as our mitigation.
Armed autonomous machines are going to be available this decade in numbers. Our enemies will certainly deploy them.
Last edited:
You have a good point but the genie has already exited the bottle. AI controlled drones are already here.
.
In a war scenario where your country’s continued existence, is under threat, you aren’t going to nerf your drone capability. Swarm attacks are the new buzzword and that requires AI with a human operator hopefully sitting on the loop somewhere overseeing the operation.
Of course to add to the complexity your enemy is jamming communications between operators and drones which gives you two options. You either accept a mission kill or you allow AI to complete the mission for you.
War is a dirty business. Soon we will see battlefields swarming with AI controlled killer drones hunting down soldiers and attacking anything it deems to be a target.
What concerns me more is this technology will be available to any bad actor.
Last edited:
Todjaeger
Potstirrer
One, there is the question of whether or not Aegis (and more specifically, which base lines) have an autonomous mode. Two, there are the related questions of what is, or what does the autonomous mode cover? I rather doubt that an Aegis system used by the USN would autonomously be able to launch TLAM's. VL-ASROC IMO would also be an unlikely candidate for autonomous launch, simply because of how a potential sub contact would likely need to be prosecuted.
Where an Aegis autonomous mode might make some sense would be in a self-defence mode, where the CMS might start automatically cueing inbounds ID'd as hostiles for engagement, and then possibly oversea SAM launches for the intercepts.
OTOH with CEC, the potential information situation can become much more saturated and squiffy. AI could be used to help sort through the data to collect and present it to system operators in a more digestible and relevant data package.
Aegis has no automated capability to launch Tomahawk. In almost every engagement of any sort command approval is required. Autonomous defence actions can be set, but the engagement parameters are strictly limited, usually to such things as countermeasures. There is certainly no world wide Aegis. Aegis platforms communicate through CEC and the various link systems, but these are manually controlled and limited in area.