Australian government policy is to replace the current six Collins class with twelve new larger submarines armed with, among other things, long-range cruise missiles. The hope is that these craft will provide something like the long-range punch formerly provided by the recently-retired F-111s. Where is submarine technology going? How may it affect any new Australian submarines over, say, the next two decades? Any discussion should begin with the reasons that submarines, albeit expensive, are still worthwhile. The fundamental value of submarines is that they are stealthy, hence can operate in areas nominally dominated by others. What the submarine does with that ability varies with what it has on board and with the scenario. In the past, the most prominent missions have been to attack enemy ships and submarines, to conduct reconnaissance, and to deter an enemy by threatening strategic attack from a secure place. (KYM TO TASHA, BPS) Australia does not currently entertain the vision of strategic deterrence by submarine, but that might become attractive in a future in which nuclear weapons were more widely spread in the region.

22nd Dec 2010

Australian government policy is to replace the current six Collins class with twelve new larger submarines armed with, among other things, long-range cruise missiles. The hope is that these craft will provide something like the long-range punch formerly provided by the recently-retired F-111s. Where is submarine technology going? How may it affect any new Australian submarines over, say, the next two decades? Any discussion should begin with the reasons that submarines, albeit expensive, are still worthwhile. The fundamental value of submarines is that they are stealthy, hence can operate in areas nominally dominated by others. What the submarine does with that ability varies with what it has on board and with the scenario. In the past, the most prominent missions have been to attack enemy ships and submarines, to conduct reconnaissance, and to deter an enemy by threatening strategic attack from a secure place. (KYM TO TASHA, BPS) Australia does not currently entertain the vision of strategic deterrence by submarine, but that might become attractive in a future in which nuclear weapons were more widely spread in the region.

Any new Australian design will probably begin in the next few years, and thus will draw on technology developed since the Collins class was built. The most dramatic changes have probably been in combat system technology and in propulsion.


Australian submarines tend to be large because they have to operate for extended periods far from home. They need the endurance to transit over those distances, and the ability to operate quietly for an extended period once they reach their operating areas. Because their mission is often probably surveillance of distant areas, nothing in current technology is likely to make long range less necessary or to reduce the endurance they need once they arrive. Like other submarines, they will normally operate at low speed, largely because no existing or likely propulsion system (other than nuclear) will give them a combination of high sustained speed and long endurance. However, they need high burst speed to escape if they are detected. Whether the current high speed time limit for non-nuclear submarines, about an hour, is enough depends on how effective enemy sensors become.

When the Collins class was designed, there was considerable interest in air-independent propulsion (AIP), but it was all speculative, with alternative systems proposed but by no means in service. About a quarter-century later the German Navy routinely operates its fuel cell system and the Royal Swedish Navy has run a submarine from Sweden to the Mediterranean entirely on air-independent power, using a Stirling engine. (KYM TO TASHA, BPS) Both navies claim that a submarine operating with AIP is orders of magnitude quieter than one snorkelling, hence is virtually immune to passive detection (others may disagree).

Current forms of AIP offer substantial engine endurance at low speed, and hence are well-adapted to a submarine attempting to loiter quietly in an operational area, e.g. one making a strategic patrol. The great question has always been whether a next-generation form of AIP, such as a fuel cell, could replace both submarine diesels and batteries altogether. In the past the answer, based on the energy content of the fuel, has been that a submarine relying entirely on AIP could operate at moderate speed for several weeks, but not for the sort of patrols the Royal Australian Navy envisages. AIP is not, and cannot be, equivalent to nuclear propulsion. How valuable the RAN will find AIP should, it seems, depend on how effectively its likely future enemies can conduct passive underwater surveillance. The better their surveillance, the more the RAN may come to believe that any alternative to AIP would endanger submarines conducting their own patrols in distant waters. However, it seems unlikely that the RAN can replace conventional propulsion with AIP, because AIP does not give the sort of endurance it wants, or the sort of high-speed power which saves a submarine once it is being pursued.

Only nuclear power can give that; a nuclear submarine can be silenced effectively (at a substantial cost in size and complexity, it is true) but also enjoys high underwater performance. If Argentina, Brazil, and India succeed in building nuclear submarines, as they currently claim they will do, the RAN may become substantially more interested in nuclear propulsion for the future, given the vast distances its submarines must traverse. (KYM TO TASHA, BPS)


One possibility not widely discussed in recent years is that a submarine can gain speed by using polymers to smooth water flow over her hull. Polymers were often discussed in the past, but attempts to use them failed, partly because impurities in the water flowing over the submarine tended to overcome their effects. It is not clear whether a future submarine might benefit from revived interest in polymers or from some other flow-smoothing mechanism, such as a new outer hull material. Because the resistance to a submarine's motion through the water comes from almost entirely from friction of the water flowing over her hull, anything that drastically reduced that friction might have an enormous effect. For example, it might greatly increase the time at maximum speed by cutting the power required to maintain that speed.

As an important aside, nuclear power provides a submarine with substantial reserves of electric power, something not available from her batteries or from an AIP system which can barely drive the submarine. Electric power reserves may become much more important if the character of submarine warfare changes, as the rest of this article will suggest.


Submarines are increasingly expensive; their unit costs are at least comparable to those of well-armed frigates and destroyers. Submarine numbers have therefore fallen in most navies. The question then is whether the footprint of an individual submarine can somehow be expanded, so that she does more than in the past. In the anti-submarine role, that has usually meant better passive sensing, for longer sensing ranges -- but then the question is whether the submarine can exploit what it hears. For example, the Collins class has vastly better passive sensors than predecessors, and passive towed arrays in particular have seen considerable improvement.

In an Australian anti-submarine context, extended submarine sensing performance would fit most logically with the use of maritime patrol aircraft to run down distant contacts, the key issue being communication between the submarine and its base (the base would connect to the aircraft system). Attempts at vastly-expanded submarine anti-submarine weapon range, achievable by substituting an underwater-launch missile for the torpedo, have generally been abandoned because anything short of a nuclear warhead is unlikely to make up for the imprecision of the distant location datum (the U.S. Navy weapon in this class was Subroc, and a non-nuclear [torpedo-armed] version was considered but dropped). One reason that passive sensing might not infinitely improve Australian submarine anti-submarine performance is that many of the likely targets are diesel-electric submarines with small (or potentially small) signatures.

By way of contrast, nuclear submarines offer good opportunities for passive sensing because their machinery produces consistent, even if low, noises, and, at least in theory, any consistent sound can be detected if it persists long enough against the random background of the sea. Thus the future of this type of detector in the RAN depends on what targets are envisaged. Submarine sensors of course detect surface ships, and in general surface ships are considerably noisier than submarines.

Unmanned Underwater Vehicles.

The single most dramatic combat system change -- and the one most likely to affect future submarines -- since the advent of the Collins class has been the rise of unmanned underwater vehicles (UUVs). They offer a different kind of extended footprint. A crude example might be a submarine watching a port, assigned to follow (or attack) a particular emerging surface ship. If the port has several entrances, what is the submarine to do? If the submarine can maintain UUVs in the different entrances, and maintain underwater communication with them, the situation is transformed; the submarine's presence is multiplied. As of now, no one has solved the problem of extended UUV endurance; existing batteries and fuel cells do not deliver enough energy to provide sufficient endurance, even at low speed, or to keep powering sensors. (KYM TO TASHA, BPS) That is, it is very much easier to power a fast torpedo for, say, thirty minutes, than to power a much slower torpedo-sized UUV for say thirty hours.

The UUV also needs sufficient internal intelligence to deal with contingencies such as underwater obstacles. The situation for underwater communications is better. Since the late 1990s it has been possible to communicate effectively over substantial distances, generally in digital form; it has even been possible to pass periscope pictures. This new capability inspired the then U.S. Naval Underwater Weapons Center at New London to propose a new kind of UUV, Manta. Manta was armed with torpedoes, and enjoyed both long endurance and high maximum speed. NUWC circulated a video in which a U.S. submarine (ca 2050, to solve the energy problem) blockaded a port with two entrances using a pair of Mantas. One of the Mantas detected the target, provided enough information to insure that it was the right target, and was ordered to engage it. In effect this is an underwater equivalent of what is happening in the aircraft world, except that the potential numbers are smaller and ranges are much shorter. In NUWC's vision, Manta was carried conformally on the nose of a future attack submarine (four could fit) and the Mantas were, in effect, the submarine's torpedo tubes. They could be reloaded from inside the submarine. NUWC built a model of Manta, but the technology, particularly that for propulsion, does not yet exist. However, anyone looking at the future into which the next-generation RAN submarines would fit would have to take something like Manta into account as a highly-desirable feature.

Current UUVs are far more primitive, but they still present interesting possibilities. About 2001 a U.S. Navy group considering the future of submarines built on the fact that a submarine was extraordinarily well adapted to approach a potentially hostile coast to collect key information prior to the approach of the fleet. For example, it could launch one or more UUVs to conduct mine reconnaissance. Given how difficult it is to be sure that a minefield has been neutralized, reconnaissance may be the way to evade mines. It sidesteps the usual problem of mine hunting, which is to decide whether an unknown object really is a mine. Mine reconnaissance UUVs are currently in service in the U.S. Navy.


Submarines increasingly perform a vital electronic reconnaissance role. All other forms of electronic reconnaissance are more overt, and an enemy can choose to shut down transmissions while an airplane or satellite is within range (hence the enormous efforts to conceal which aircraft and which satellites are collecting data). A submarine officer once pointed out that U.S. submarines on such duty picked up less than 10 percent of transmissions, but these were often the really significant 10 percent. However, a single submarine can collect in only a limited area. If the submarine operated a group of UUVs, each capable of detecting emissions, then her footprint might be considerably expanded. (KYM TO TASHA, BPS) The flock of UUVs would not be the same as the submarine, because collection includes judgments by the operators as to which signals are worth collecting. Conceivably an underwater link to the collecting UUVs would help, but that would not be the same thing.

Even so, the expanded footprint would be extremely valuable. The submarine would function as mother ship, and would operate further offshore -- where she would be freer to transmit what she collected back home. To the extent that electronic reconnaissance is often the best source of information in a crisis, the ability to do so on an extended basis, at a limited cost in submarine hulls, would seem to be particularly attractive to the Australian government. The U.S. EDO company exhibited an electronic reconnaissance system specifically for UUVs a few years ago, but it seems to have been a proposal rather than a serious project at the time. Even so, this would seem to be a vital UUV role.

UUVs are also attractive for anti-submarine warfare -- if the energy problem can be solved. They are clearly far more covert than surface ships or even normal submarines, because they are so small. Many small countries operate limited numbers of submarines, and all of them have limited endurance (as, indeed, do Australian diesel-electric submarines). These craft must go back to their bases periodically for refuelling and replenishment. Dealing with them near the bases is a time-honored anti-submarine tactic. In an extended sense, it was the tactic NATO and the U.S. Navy planned to use against the Soviets, by blocking passages such as that near the Aleutians from the Soviet base at Petropavlovsk. The idea was always that, over time, Soviet submarines which survived their patrols would still have to return to base, hence pass through the barriers. Obviously the Soviets planned to fight at the barriers.

A covert UUV capable of getting into the enemy's harbor and following the submarine back out to sea would be quite valuable. At the least, it could be commanded to beacon other forces to attack the submarine. Even if the energy problem remains difficult, one can imagine a UUV entering an enemy's submarine base and tagging his submarines with transponders, which would be used to reveal the submarines at sea (the transponders would give the submarines considerable signatures, in effect, denying them the value of their built-in stealth).

Special Forces

UUVs might be associated with the other kind of vehicle submarines sometimes launch, the swimmer delivery vehicle for special forces. Right now most special force swimmers simply exit through a torpedo tube or a hatch, but that limits the distance the submarine can stand off from the beach of interest. In that sense the swimmer delivery vehicle extends the submarine's footprint, since it can reach out to a wider area. The U.S. Navy has experimented with UUVs as means of transportation back and forth to a deployed group of special forces, for example to keep them supplied. (KYM TO TASHA, BPS) In an exercise, a UUV transported samples from the deployed special force back to the parent submarine, where they were analyzed (this exercise simulated an operation against a possible weapons site).

The parent submarine may also operate unmanned air vehicles (UAVs); there is considerable current interest in submarine-launched UAVs which can be recovered underwater. Such a device is half UAV and half UUV, because once in the water it has to navigate to the submerged submarine, perhaps on the basis of a sonar beacon. The device might be launched from a torpedo tube, navigate to the surface, and then fly off. For example, a submarine might launch a UAV which would scout for the special forces unit the submarine was intended to launch. A UAV might also be used to target land-attack missiles the submarine carried, perhaps to neutralize potential opposition to the small special-forces unit.

Impact of UUVs

What does all of this mean for the submarine itself? Many current UUVs are similar in shape and size to torpedoes. Submarines have limited internal volume, so each displaces a torpedo. The U.S. Navy developed a UUV launch-and-retrieval system which fit two of a nuclear submarine's four torpedo tubes -- halving the submarine's instant weapon-launch capacity. In the U.S. case, that may have been acceptable because the submarine also had twelve vertical tubes for Tomahawk missiles, but it certainly badly reduced the submarine's torpedo capacity. The vertical tubes are less attractive for UUV use because there is no direct access from inside the submarine, e.g. to service the UUV.

Overall, if UUVs (and perhaps also UAVs) are the most important new development for submarines, the implication seems to be that future submarines need more internal volume and more tube space (or tube-like space) for launch and recovery. Some concepts for future submarines show multiple hulls, for example two hulls side by side, to provide that extra volume without requiring excessive hull diameter or draft. The envelope built around the multiple hulls might provide space for vertical tubes, for example for missile and UAV launch.

The other implication is internal. UUVs help change the character of a submarine's combat system. In a current submarine like HMAS Collins, the combat system is conceived largely to create a valid underwater picture from the ship's sensors (mainly sonars, passive and active). The ship's command fights the ship based on that picture. For modern systems, the greatest challenge is to turn passive sonar data (which gives the direction to a sound source) into a full range and direction picture, based on the way in which direction seems to change as the submarine moves.

How quickly the picture can be created depends partly on how fast the underlying computers are, and also on how complex the processing can be. U.S. experience, reflected in the system currently on board the Collins class, has been that it is far more efficient to improve processing than to bet on more sensitive sensors (work is proceeding on lighter-weight sensors, however). Modern submarine command systems vary in the extent to which operators can keep forming the tactical picture while an attack is planned and executed. The larger the command space, the more consoles it can accommodate, hence the more separate functions it can more or less simultaneously conduct. (KYM TO TASHA, BPS) As submarines operate more often in littoral areas, they encounter more and more discrete targets (surface ships, submarines, false targets due to noise), and simply disentangling the tactical picture becomes a more elaborate exercise. Conversely, increassed computing power can make it possible for the automated part of the submarine combat system to take over more of this load, as in current U.S. submarines and, presumably, the Collins class.

UUVs and UAVs (and optronic periscopes) all add to the processing burden. Again, U.S. experience has been that it is best to pipe all available sensor data into an integrated combat system, rather than split it up between different sources of data (e.g., ship's own sonar separated clearly from, say, passive electronic intercepts via the periscope). That was done in the combat system on board the current Virginia class. To some extent the new idea of a completely integrated combat system reflects increased U.S. interest in littoral warfare, in areas in which electronic intercepts may provide better information than passive sonar. However, the new way of doing submarine business seems inescapable if it turns out that the submarine is best seen as mother ship for unmanned vehicles (and even for manned vehicles carrying Special Forces). The U.S. Navy is unusual in splitting picture formation from fire control; normally the same set of command consoles does both. The reason was interesting. As electronics rapidly evolved, it paid to insert new elements into the picture-keeping part of the system. However, anything which could launch a weapon required elaborate certification -- as was quite proper. If the two functions were integrated tightly together, the entire system would have to be recertified quite regularly. If they were split, recertification could be limited to the weapons control element.


The other important new U.S. development is A-RCI, Acoustic Rapid COTS Integration. A-RCI began in the late 1990s, when it was clear that U.S. submarines were losing their acoustic advantage, but also that there was no money for new specially-designed sensors or, for that matter, combat systems. It was also clear that commercial electronic systems were advancing rapidly: that is usually expressed by Moore's Law, typically rendered as the statement that capability doubles every 18 months. Somehow submarine combat systems had to be opened up to exploit commercial developments. A-RCI installed a fiber-optic bus which could carry sensor data around a network of processors. The navy concentrated not on the 'state of the art,' but on periodic changes of technology to reflect the 'state of the practice,' i.e., what was reasonably reliable -- to the point where processors typically do not need to be serviced during a submarine's deployment. New processors made it possible to deploy new kinds of signal processing.

The great surprise of A-RCI was that submarine capabilities increased dramatically even though submarine sensors did not change. (KYM TO TASHA, BPS) Clearly, better sensors would have added performance - but existing ones were hardly being fully utilized. It was far more efficient, and far less expensive, to concentrate on electronics, which was relatively easy to refresh. A-RCI proved so successful that similar initiatives were taken on board surface ships and in naval aircraft. It seems clear that without something like A-RCI it will be impossible to exploit rapid improvements in new kinds of systems such as UUVs and submarine-launched UAVs. However, without such means of extending the submarine's footprint, submarines may be pricing themselves out of affordability. A-RCI was the occasion for splitting submarine combat control between picture-keeping and weapons control.


Greater processing power makes it possible to rethink the submarine's weapons, too. A torpedo is, in effect, a one-way UUV. The Germans already use their torpedo seekers as acoustic probes; they are wired into combat systems to act as mobile sonars. As long as it can communicate back to the submarine, the torpedo can add to the submarine's underwater picutre, and the submarine can help the torpedo distinguish real from false targets. On this basis it pays to be able to order the torpedo to loiter. The torpedo might then be held near the potential target until the commanding submarine decided which of several possible targets was worth attacking. Current guidance wires hardly seem suited to this sort of operation, but the advent of effective underwater acoustic communication suggests greater possibilities. No current torpedo seems to have been designed to realize this potential, but this possibility has been discussed by Saab, which makes Swedish torpedoes. Similarly, a minefield laid by a submarine could also be used as a sensor field, the submarine interrogating it periodically via long-range acoustic links.


Overall, it seems likely that future submarines will be considerably larger than their current equivalents, because otherwise their costs will seem disproportionate. It takes size to accommodate footprint-extenders like UUVs and submarine-launched UAVs -- not only to carry and launch them, but also to service them and to exploit the information they retrieve. It will take more onboard power to handle the larger combat systems fed by such sensors (the power to move a larger submarine does not grow as rapidly as other demands). These considerations would seem to make nuclear power considerably more attractive, ultimately, than diesel-electric or some form of AIP propulsion, simply because the latter work best for a submarine with minimal power demands -- not rapidly-growing ones. The enormous potential of a largely invisible warship would seem to guarantee the submarine's future -- but not the form that future will take.

* All opinions expressed in this article are the author's own, and do not necessarily reflect those of the U.S. Navy or of any other organization with which he has been associated.

APDR at a glance