Potential Missions for Future PLA Unmanned Aerial Vehicles

I recently came across a 2013 Project 2049 Institute monograph detailing PLA efforts to research and develop UAV technologies. Ian Easton’s and Russell Hsiao’s report pieces together the PLA organizations, academic institutions, and industrial activities involved in Chinese UAV work; this is no small open source achievement. More importantly, though, it taps Chinese-language sources to outline concepts from each of the PLA’s services regarding potential future uses for UAVs. Many of these concepts unsurprisingly mirror a number of those under consideration by the U.S. armed services:

• Long-range autonomous strike
• “Wingman” duties for manned aircraft
• Localized communications relay
• Anti-ship scouting and targeting
• Ground combat scouting and targeting
• Wide-area surveillance

They make an additional key observation regarding the possibility that expanded PLA UAV capabilities might incentivize increased Chinese brinksmanship, and possibly the use of force, in a crisis:

“There could be a sense that because human pilot lives are not at stake, operators can push farther than they otherwise might. It is also not clear how nations would react to isolated UAV attacks in times of crisis, especially if these were blamed on mechanical or technical failure, or even on cyber hackers. In the future, PRC decision-makers might feel compelled to order “plausibly deniable” UAV attacks as a means of sending a political signal only to inadvertently wind up escalating tensions.” (Pg 13)

This dovetails closely to some of my own observations on unmanned systems and escalation management. The main difference is that whereas I proposed that an opponent’s unmanned scouts should be considered fair game for attacks during a crisis depending upon the circumstances at hand, it is entirely possible that an opponent might go further and use its unmanned vehicles to conduct limited attacks on traditional targets for coercive effect. The authors don’t argue that the PLA is considering use of UAVs for this kind of purpose, but they are correct that the PLA or any other UAV-operating military might.  The implications for crisis management deserve systematic examination through war-gaming.

Some of their most interesting but in no way surprising observations concern Chinese writings regarding the potential uses of UAVs to support anti-ship attacks. One such use proposed in the source writings is for UAVs to simulate inbound raiders, with the intent being to lure an opponent’s screening aircraft and surface combatants into wasting long-range anti-air missiles against these decoys. Other UAVs might perform electronic attacks against radars and communications systems. All this represents a longstanding and well-understood set of tactics. The requisite technical, tactical, and doctrinal countermeasures are similarly well-understood: multi-phenomenology outer-layer sensors that can classify contacts with high confidence, robust combat training to psychologically condition crews for the possibility of hostile deception, deep defensive ordnance inventories, and embracing tactical flexibility/seizing the tactical initiative. The only question is the defender’s will to invest in these kinds of countermeasures—both materially and culturally.

Easton and Hsiao also note that Chinese writers have proposed that some UAVs might perform direct ‘suicidal’ attacks against radars or warships (and in doing so fully blur the line between UAV and cruise missile). The Chinese sources additionally suggest that UAVs could replace manned aircraft as anti-ship missile-armed raiders, though I would argue this presumes the requisite artificial intelligence technologies for conducting attacks against ‘uncooperative’ targets in an ambiguous and dynamic tactical environment reach maturity.

Lastly, Easton and Hsiao’s sources suggest UAVs could serve as communications relay nodes that support anti-ship attacker—and perhaps in-flight missiles as well. For example, a scout UAV could conceivably provide targeting-quality cues to an over-the-horizon “shooter” via a relay UAV, and then provide periodic targeting data updates to the in-flight missiles thereafter. Or the relay UAV might enable direct communications between “shooters” within a given area. It might even enable direct coordination between in-flight missiles approaching on different axes. The use of highly-directional line-of-sight communications pathways or low probability of intercept transmission techniques would make this a particularly vexing threat. Clearly, naval battleforces will need means of detecting and classifying relay UAVs (not to mention scout UAVs) lurking in their vicinity.  

Easton and Hsiao observe that even though the sources they reviewed for their monograph wrote relatively little about using UAVs in the aforementioned ways for land-attack or ground warfare missions, there are no fundamental factors that prevent them from being extensible beyond the anti-ship mission. They’re absolutely correct on that point, and that’s something that all the U.S. armed services should be thinking about for the future.

--Updated 7/16/15 10:54PM EDT to fix first link in post--

The views expressed herein are solely those of the author and are presented in his personal capacity. They do not reflect the official positions of Systems Planning and Analysis, and to the author’s knowledge do not reflect the policies or positions of the U.S. Department of Defense, any U.S. armed service, or any other U.S. Government agency.

A Weapon’s Range is not the Whole Story

Illustration of U.S. and threat anti-ship missile ranges from Bryan Clark's CSBA monograph "Commanding the Seas: A Plan to Reinvigorate U.S. Navy Surface Warfare" (pg 13).

The U.S. Navy is clearly at a deficit relative to its competitors regarding anti-ship missile range. This is thankfully changing regardless of whether we’re talking about LRASM, a Tomahawk-derived system, or other possible solutions.

It should be noted, though, that a weapon’s range on its own is not a sufficient measure of its utility. This is especially important when comparing our arsenal to those possessed by potential adversaries. A weapon cannot be evaluated outside the context of the surveillance and reconnaissance apparatus that supports its employment and the overall size of its inventory.

Here’s an example I set up in the endnotes of my article on maritime deception and concealment regarding effective first strike/salvo range at the opening of a conflict:

Optimal first-strike range is not necessarily the same as the maximum physical reach of the longest-ranged weapon system effective against a given target type (i.e., the combined range of the firing platform and the weapon it carries). Rather, it is defined by trade-offs in surveillance and reconnaissance effectiveness and in the number of weapons employable in a short time as the target’s distance from the firing platform’s starting position increases. This means that a potential adversary with a weapon system that can reach distance D from the homeland’s border but can achieve timely and high-confidence peacetime cueing or targeting only within a radius of 0.75D has an optimal first-strike range of 0.75D. It follows that if, for technical, operational, or logistical reasons, the adversary can fire only a few D-range weapons within a defined short period of time, and if his doctrine therefore calls for using D-range weapons in coordination with far more plentiful weapons of range 0.5D, the optimal first-strike range decreases to 0.5D. This does not reduce the dangers faced by the defender at distance D but does offer more flexibility in using force-level doctrine, posture, plans, and capabilities to manage risks.(Pg. 113-114)

The same logic applies following the first strike/salvo, except that the victim of that attack will enjoy much more relaxed rules of engagement for countering the opponent’s surveillance and reconnaissance efforts. In this case, let’s say that Blue possesses a long-range weapon with maximum physical reach of R. Let’s also say that both sides’ anti-scouting efforts inside a given area have neutralized or destroyed surveillance and reconnaissance sensors, not to mention disrupted or degraded sensor-to-network connectivity. Let’s stipulate that Blue’s fraction of R that is covered by high confidence surveillance and reconnaissance is x, and Red’s equivalent fraction of D is y. All other factors being equal, the basic advantage consequently goes to which multiple is greater: Blue’s xR or Red’s yD. Note that if either side's confident surveillance/reconnaissance coverage exceeds their weapons' maximum physical ranges, then that full range can be realized (x or y = 1.0).

The qualities and quantities of sensors, and the architecture and counter-detectability of the (electromagnetic through the air and space, acoustic under the water) data pathways they use to relay their measurements to ‘consumers’ matter just as much as weapon range. Under intense anti-scouting opposition, they arguably matter even more. Remember Wayne Hughes’s maxim: attack effectively first.

Now throw in the inventory-size and salvo-rate considerations on top of the sensing competition. Assuming inventory survivability against attacks is held equal for both sides, if one belligerent has many times more of its long-range weapon than the other belligerent has of its long-range weapon, the former may gain considerable campaign-level advantages over the latter such as greater operational flexibility or greater tolerance for taking operational risks. If one belligerent can salvo off more of its long-range weapon within a short period of time than the other can, then obvious tactical advantages can accrue there as well.

The bottom line is that the question of striking clout is far more complicated than a comparison of range alone. Surveillance/reconnaissance quality and weapons inventory quantity are just as important. The full picture must always be considered.

--Updated 2/26/15 10:32PM based on comments--

The views expressed herein are solely those of the author and are presented in his personal capacity. They do not reflect the official positions of Systems Planning and Analysis, and to the author’s knowledge do not reflect the policies or positions of the U.S. Department of Defense, any U.S. armed service, or any other U.S. Government agency.  
 

Omniscient, Impregnable Ocean Surveillance Systems?

Pretty much every time the topic of the aircraft carrier’s future combat viability comes up, or the general viability of surface fleet operations within a contested zone for that matter, we hear arguments that potential adversaries’ surveillance systems will assuredly turn the waters they monitor into graveyards for American and allied warships. Sensors are becoming too sensitive and ubiquitous, we’re told. Data fusion hardware and software are becoming so powerful that platforms—especially large ones—will soon no longer be able to hide amidst the ‘background noise,’ we’re warned. Efforts to disrupt an adversary’s situational picture will demand cyber-penetration of his surveillance/reconnaissance networks, we’re admonished. Perhaps the boldest claim, often made implicitly, is that the quality of the adversary’s situational picture will remain pristine as a conflict wears on. And so on. I may be paraphrasing these arguments, but I am not exaggerating what has sometimes been asserted.

Now, I fully agree that the challenge posed by advancing sensor, data fusion, and networking technologies is severe. The threat is real and growing; much effort in countermeasure development and procurement, doctrine and tactics development, and force-wide training will be necessary if we are to respond effectively. There are two serious common flaws in all of the above arguments, though.

First, they are essentially asserting that maritime surveillance/reconnaissance ‘systems of systems’ are virtually omniscient and impregnable. If that is indeed true, then I must commend our potential adversaries on developing the first flawless sensing, communications, and human decision-making systems in history. Somehow I doubt they’ve reached such a milestone.

Second, they attribute capabilities to potential adversaries’ surveillance systems, skills to those systems’ operators, and genius to those operators’ commanders without acknowledging that the very same capabilities, skills, and genius can be present on our side as well. Physics doesn’t pay attention to national flags. Nor does human psychology amidst war’s fog and friction.

Just because the technical, tactical, and operational difficulties of overcoming advanced wide-area surveillance systems—and the humans who use them—would be considerable most assuredly does NOT mean that it could not be done when needed. One of the constant themes in my writings has been to point out the things the U.S. Navy did during the Cold War against a then-state-of-the-art ocean surveillance system and a highly intelligent opponent. Again, while sensor, processor, and communications technologies have advanced, so have the technologies and techniques available for use as countermeasures. It’s part of the never-ending, iterative struggle between offense and defense, scouting and anti-scouting. This struggle is as old as war itself. Sometimes offense or the scout possesses the upper-hand, and at some point thereafter it is captured by the defense or the anti-scout. The switching of who holds the advantage can take place extraordinarily quickly; one should read Martin Bollinger’s excellent book on the WWII competition between German radio-controlled glide bomb developers and allied Electronic Warfare (EW) countermeasure developers, Warriors and Wizards, to see how this occurred in practice. Likewise, one should read Norman Friedman’s seminal explorations of the Cold War competition between Soviet maritime scouting and U.S. Navy anti-scouting. While it is true that these U.S. Navy capabilities were never tested in war against the Soviets, the very same is true about the Soviets' capabilities.

I fully agree that we mustn’t assume anything about our abilities at any given point in time to degrade any particular potential adversary’s surveillance/reconnaissance picture. I also recognize that there are likely to be some key doctrinal or decision-making process differences between how the Soviet Ocean Surveillance System operated and how our contemporary potential adversaries’ ocean surveillance systems operate. Nevertheless, I vehemently disagree with crediting our potential adversaries with abilities to track and target our forces seamlessly while outright denying us any possibilities for disrupting or degrading that coverage. After all, the underlying technologies, tactical principles, and physics/psychology-based exploitable vulnerabilities are largely available to both sides.

As for the assumption that we need to cyber-penetrate adversary networks if we are to handicap them, well, that neglects the many possibilities of messing with their situational pictures more directly. Restoration of our Cold War-era ability to conduct complex tasks, including protracted carrier air operations, under highly restrictive Emissions Control conditions. Conduct electronic or kinetic attacks as possible to disrupt, distract, or destroy their sensors and electromagnetic/acoustic data relay paths located on, under, or over the ocean. Use less-valuable platforms to simulate more valuable platforms. Saturate their picture with real contacts, of which only a few are the platforms requiring protection. Overwhelm them with multiple lines of operations, of which only one or two are ‘real’ and the others feints or demonstrations. Induce them to prematurely burn up their limited inventories of their longest-range weapon systems. Show them something they expect to see, and then exploit their reactions. Many combinations of these options can be tailored to meet a particular situation.

Yes, adversary systems can have strong counter-countermeasures. No system is perfect, however, and sometimes the best counter-surveillance and counter-targeting measures are more psychological than technological. It is terrific when we possess enough intelligence on the inner workings of some threat system to blind or deceive it through our clever countervailing use of technology. That said, blinding and deception can be accomplished with just as much impact by presenting the adversary with a misleading ‘picture’ consisting of actual platforms, emissions, and the like to distract him, confuse him, or perhaps even impel him to take actions we can exploit.

It needs to be appreciated that the ocean surveillance systems under discussion are a threat to any naval combatant, not just aircraft carriers. Even U.S. submarines will face an undersea equivalent—albeit with less area under effective surveillance—in the future. These systems must be suppressed, deceived, or attrited, whether temporarily or permanently, if naval operations within a given segment of a contested zone are to be conducted at a tolerable level of risk and with a tolerable allowance for damaged or lost platforms.

It follows that the sequence of operations in a campaign, and the overall design of that campaign, will matter immensely. Why does a carrier (or any other surface platform with medium to high campaign-value) have to ‘go downtown’ immediately at the start of a conflict? We certainly didn’t do that in the Second World War; our prewar plans eventually came to rule out such a high-risk “through-ticket” approach. Despite many assumptions then and now to the contrary, even the Maritime Strategy of the 1980s didn’t demand surface operations that far forward early in a Soviet-American war. If we want to telegraph to a contemporary potential adversary that any conflict would assuredly not be short, cheap, or low-risk, perhaps designing operating concepts that have a strong chance of ensuring conflict protraction would be best for strengthening our conventional deterrence credibility? The beauty of having many combat arms across our four services—not to mention those of our ‘frontline’ allies who would be inherent parties to many of the highest-danger future conflicts we face—is that it creates campaign design options. The successful use of precursor or parallel operations to pave the way for temporary localized U.S. scouting/anti-scouting superiority in a main operation or its successor operations is quite possible even in a major peer-level conflict.

The bottom line is that we have access to a much larger toolbox than many are willing to acknowledge. I am not arguing that deception and concealment are silver bullets that can assuredly shield our forces under all circumstances. Hardly so; like any of the other tools for gaining operational access, they could fail in action. I certainly wouldn’t build a strategy, campaign plan, or operating concept around an assumption that our deception and concealment efforts would be successful. However, when their intelligent use is balanced with imaginatively-structured operations rooted in calculated risk (including the design of branching actions that account for failed counter-surveillance), and with a LOT of training and discipline, a LOT of options and opportunities open up. In fact, some potential adversaries’ network-dependent architectures and C² doctrine for performing maritime denial tasks are just as—if not more—vulnerable to what I’ve outlined than our force-level networking architectures and C² doctrine are…or have to be. Operations on interior lines of networking combined with decentralized C² doctrine that embraces command-by-negation are simply essential to us in that regard.

We learned to do counter-surveillance and counter-targeting extremely well during the Cold War. I don’t see why we can’t—with the requisite focus and investments—get back to that level. I believe that the debates over future naval force architecture and doctrine, not to mention theater strategy in general, need to better address these questions. Reasonable people can disagree with the arguments I've made, but I would welcome counterarguments that are rooted in theory, historical case studies, or empirical assessments of surveillance and countersurveillance technologies' inherent capabilities AND limitations. It's time to stop overlooking ‘the dark arts,’ or otherwise dismissing them out of hand. It's time to stop unquestioningly awarding our potential adversaries an invincible degree of surveillance dominance on paper.

The views expressed herein are solely those of the author and are presented in his personal capacity. They do not reflect the official positions of Systems Planning and Analysis, and to the author’s knowledge do not reflect the policies or positions of the U.S. Department of Defense, any U.S. armed service, or any other U.S. Government agency.

 

The Coming (Invisible) War of the UUVs?

In my post earlier this week, I talked about how the neutralization or destruction of unmanned scouts during crises might not be terribly escalatory.

Following that post, a colleague pointed out to me some important and subtle differences between how anti-scouting unfolds at and above the surface versus how it unfolds underwater. The beauty of undersea operations is, of course, that their effects are almost always plausibly deniable. This is the case today with submarines, and is sure to be the case with underwater unmanned vehicles.

Nevertheless, while a submarine that fails to come home cannot be covered up, the same will not be true with respect to their future unmanned brethren. Indeed, the fact that unmanned underwater operations will generally not be publicly disclosed means that even if a country employing these vehicles realizes that one or more were lost due to hostile action during a crisis, it is extremely unlikely that its leaders would be willing to escalate outside the unmanned underwater realm. Nor would it be easy for the victim to make a strong (and publicly-releasable) case attributing the loss of an Unmanned Underwater Vehicle (UUV) to a hostile act by a specific actor.

Much like history’s many invisible peacetime ‘wars’ between countries’ intelligence agencies, it seems quite likely that there will be an continuous invisible peacetime conflict in which countries boldly employ UUVs within their opponents’ territorial seas. As detailed in works such as Blind Man’s Bluff, we saw a preview of this to some extent with respect to how special-purpose submarines were used during the Cold War. UUVs will only increase these kinds of operations’ breadth, audacity, and maybe even frequency. An even greater change will be the introduction of parallel efforts to neutralize, capture, or destroy opponents’ UUVs that enter one’s own seas. UUV proliferation might lead to the creation of ‘undersea defense identification zones’ and surveillance sensor networks that support them, with countries publicly asserting the right to interdict others’ UUVs that enter these zones. Ironically, the very creation of these networks will only encourage countries to use UUVs to map, probe, closely inspect, and even meddle with the networks’ sensor arrays and communications paths during peacetime.

All the same, countries could enforce—or trespass within—these zones with minimal escalatory risk. The greatest strategic risk would be the public and diplomatic embarrassment resulting from the capture and open display of a UUV whose nationality was clearly marked or otherwise unmistakable. Perhaps, then, the most critical defensive capability installed in a UUV will be its ability to recognize it has been discovered, break back out into international waters, and break contact with its hunters. It certainly won’t hurt to design unmarked UUVs with difficult-to-attribute technical origins for use in the most audacious of missions, either.

The views expressed herein are solely those of the author and are presented in his personal capacity. They do not reflect the official positions of Systems Planning and Analysis, and to the author’s knowledge do not reflect the policies or positions of the U.S. Department of Defense, any U.S. armed service, or any other U.S. Government agency.

An Observation on Unmanned Systems and Escalation Management

One of the issues I looked at in my Naval War College Review article on maritime deception and concealment concerned an opponent’s use of unmanned scout vehicles to search for and target a defender’s platforms during a crisis. Any scout that is able to locate and then support classification of one of the defender’s individual units such as a warship (or perhaps one of the defender’s larger formations such as a naval battleforce) can hypothetically provide the opponent’s long-range strike systems with targeting-quality track data. Depending on the circumstances and the opponent’s political objectives, this targeting data might create further incentives on the margins for the opponent to execute a first strike/salvo attack.

Unmanned scouts are ideal for performing the prewar ‘tattletale’ role that until now could only be accomplished by manned platforms. These vehicles might allow the opponent to persistently cover a wider area or track a particular contact longer than might be possible if only manned scouts were used. They would obviously also avoid exposing manned scouts to interception by the defender. 

It is highly likely that U.S. rules of engagement during a crisis would preclude destroying an opponent’s manned scout in the absence of high confidence indications and warning that the opponent intended an imminent attack. U.S. use of non-kinetic means to ‘neutralize’ a manned scout would likely be limited as well. It is far less clear, though, that neutralizing or destroying an unmanned scout would carry similar escalation risks or otherwise set unacceptable escalatory precedents. Consider how Russia shot down several Georgian unmanned aerial scouts prior to the 2008 war, or how Iran shot at (and possibly also commandeered) U.S. unmanned aerial scouts over the past few years. Crisis circumstances matter, but it is worth noting that none of these cases resulted in anything more than diplomatic protests by the victims. These cases actually establish a decent international precedent regarding the legitimate bounds of a country’s response to the loss of its unmanned scouts. Granted, neither Russia nor Iran nor any other potential U.S. adversary would necessarily react along the lines of how Georgia or the U.S. did. Nevertheless, far less stigma (and international scorn) attaches to killing robots than manned platforms.  

Therefore, as I noted in my article:

A defender might declare exclusion areas during a crisis within which any detected unmanned system would be neutralized; enforcement of these areas might well not precipitate drastic escalation by the other side...A potential adversary could overcome this threat by placing unmanned systems under close manned escort, but that would undercut the rationale for using unmanned, vice manned, reconnaissance systems to increase search volume and on-station time per sortie. Also, it is not clear how a potential adversary could respond without disproportionate escalation if a defender neutralized the unmanned system by close-in jamming or other nonkinetic means. The point is worth study through war gaming. (Pg. 99, 115)

Depending upon how an unmanned scout was neutralized, the opponent might not even know within an actionable period of time whether an onboard malfunction or ‘external causes’ resulted in the loss of communications with the vehicle. This could support a U.S. ability to assert plausible deniability if need be. There would be a risk, however, that the opponent might conclude U.S. forces were operating in or near the area the scout was passing through when communications with it were lost. This might entice the opponent to focus its surveillance and reconnaissance resources on that area. All the same, if the U.S. had in fact declared a sufficiently broad exclusion area for the opponent’s unmanned vehicles, an overall U.S. deception and concealment plan might allow for disruption or even destruction of the opponent’s unmanned scouts near a specific location to create an impression that ‘something of value’ was nearby even when the actual U.S. or allied units being concealed were somewhere else entirely. Of course, nothing would prevent the opponent from pursuing similar ends via similar tactics.

I would submit that if war gaming and historical case study analysis find that the crisis stability risks of attacks against unmanned scouts would be tolerable, and if the resulting legitimization of equivalent attacks against U.S. unmanned systems would be acceptable, then it might be worthwhile for American diplomacy to advance unmanned scout neutralization (or destruction if the scout is outside the opponent’s internationally-recognized sovereign boundaries) as an international norm. This would not mean that the U.S. would automatically take such steps during a particular crisis—only that U.S. political and military leaders would reserve the option of taking them. If the opponent raised his forces’ combat readiness postures in response, that would be still be a preferable—and most likely more manageable—consequence than allowing the opponent to retain a high-confidence targeting picture that could decisively tilt him towards a decision to strike first.

The irony in all this is that it may incentivize potential adversaries to go 'back to the future.' If a potential adversary expects that U.S. forces will neutralize unmanned scouts without hesitation, then he may opt to return the brink-of-war tattletale role to manned scouts under the assumptions about U.S. rules of engagement during a crisis that I mentioned earlier. So the moral of the story may very well be: don't operate high campaign-value platforms in locations during crises where they would not be able to evade contact with the opponent's manned scouts for long. Instead, preserve those platforms in locations where the opponent cannot scout effectively, or contact with any manned scouts can be quickly and enduringly broken. And remember that whatever the opponent's prewar escalation dominance advantages of using manned scouts may be, those advantages evaporate once U.S. rules of engagement are relaxed upon the outbreak of open hostilities.

The views expressed herein are solely those of the author and are presented in his personal capacity. They do not reflect the official positions of Systems Planning and Analysis, and to the author’s knowledge do not reflect the policies or positions of the U.S. Department of Defense, any U.S. armed service, or any other U.S. Government agency.