Future-proofing the Air Force’s next bomber
Purely manned or purely unmanned aircraft possess various inherent advantages and limitations. A manned aircraft can be used in contested environments where command-and-control is limited, autonomy is required, or policy restrictions exist. An unmanned aircraft has no aircrew to limit its range and endurance, nor to place at risk of loss or capture.
Optionally manned aircraft provide the best of both worlds, allowing commanders to employ force at various risk levels and to employ their aircraft and crews to their fullest capacities. But utility in the present is not the only important measure. Such platforms — and in particular, the planned Long Range Strike-Bomber (LRS-B) — represent the lowest-cost, lowest-risk path toward a future of greater platform autonomy and technical and strategic advantage.
Preparing for Progress
Today’s technology is insufficient to allow unmanned aircraft to make independent, complex judgments in an ambiguous and novel environment and so must be tethered to human judgment. But data links can be denied or deceived, and in any case, they introduce seconds of delay that can be crucial, such as when prosecuting moving targets or surviving in well-defended airspace.
In theory, some of these limitations ought to erode as unmanned technology develops. Aircraft can already take off and land by themselves. Funded programs are developing autonomous air refueling and “sense and avoid” systems, which will allow more regular flight operations in regulated airspace. The Air Force is also funding research into tougher problems, such as automated dynamic mission planning and automated target recognition, and will continue to do so, as outlined in the service’s Remotely Piloted Aircraft Flight Plan and its Technology Horizons document, which sets priorities for science and technology development. Academic centers also continue research into artificially intelligent moral reasoning that might one day allow autonomous application of rules of engagement in complex situations. Finally, the history of electronics suggests that after 2020, computational power roughly equal to that of a human brain may be available to consumers for about what we pay today for a decent laptop.
Yet the technology that would enable true autonomy for a combat aircraft remains unusually hard and elusive. There is little reason to believe key technologies such as automated mission planning and target recognition will be ready in a decade, which is about how long it takes to design, build and test a major new aircraft, such as the LRS-B. On the other hand, such an aircraft is likely to fly 20 to 40 years beyond its initial acquisition, a span that may see the emergence of key pacing technologies and perhaps even true human-equivalent general artificial intelligence.
The best option is to build future platforms “autonomy ready” — that is, so local or remote pilots could be replaced by artificial intelligence through software upgrades rather than costly hardware retrofits or new platforms. Such design for optional manning would create an aircraft able to execute all aspects of its mission at initial operating capability while accommodating the overall trend toward greater unmanned capabilities. It neither assumes that key unmanned capabilities will be ready along with the airframe and propulsion systems nor locks out the kinds of autonomic improvements likely in the following decades.
Moreover, the expansion of unmanned operations is as much a cultural problem as a technological one. By allowing human crews to ride along as unmanned aircraft grow larger and their autonomic systems more capable, optionally manned aircraft create a space to build the trust that permits change. A crew can observe the functioning of the aircraft, provide backup or correction and refine its capabilities. This allows operators to get comfortable with ever-increasing autonomy without undue risk and to mature technology ahead of policy.
Today, unmanned systems are not trusted to plan their own missions, interpret rules of engagement or decide to employ weapons and lethal force, but will this always be the case? The trends in computation and unmanned technology argue it will not. It is difficult to predict when policymakers might have confidence in these innovations, but without the ability to evaluate them under the supervision of onboard aircrew, they might never risk large, expensive unmanned aircraft.
The Cost of Crew
But at what cost, this capability? How much more money would it take to make a given platform optionally manned, and how would performance be affected? To accommodate aircrew, certain systems must be present, no matter how big the airframe. These include triple- and quadruple-redundant aircraft control and mission management systems, environmental sensors, integrated data architecture, redundant communications apertures and equipment, and a cockpit. Each of these adds to aircraft volume, size, weight and cost. To inform acquisition decisions, we must understand where such considerations affect performance and cost and where they are negligible.
Many people assume adding a cockpit to a nominally unmanned aircraft would add significant cost or compromise. This is true for aircraft of fighter size or smaller, where accommodating a pilot would reduce range and endurance and force suboptimal engine and weapons arrangements. A single-seat cockpit would certainly affect the shape of a small aircraft like the X-47B or Boeing Phantom Ray, might account for 17 to 30 percent of empty weight, and, since cost typically scales with weight, would drive the price tag up.
However, the relative impact of a cockpit diminishes as the aircraft grows to accommodate larger payload and range requirements. As estimated by the Center for Strategic and Budgetary Assessments, an aircraft in the class of LRS-B would require a payload above 10,000 pounds and a range of 5,000 nautical miles. A two-pilot cockpit would add about 6,000 pounds, representing just 4 to 6 percent of empty weight and cost. Nor would cockpit volume much affect range; in an aircraft of this size, filling the entire cockpit space with fuel would add less than 100 nautical miles, a 2 percent difference.
A cockpit also introduces other considerations: The aircrew must be able to see outside the aircraft, and it must have air to breathe, some control over its climate and a way to escape. Cockpit angle directly affects aircraft shaping, a primary characteristic of stealth properties. But for large, low-observable aircraft, the required angles to accommodate aircrew visibility and escape do not much affect shaping or engine position, which are driven primarily by the size and, especially, height of the payload bay.
The point where the cockpit no longer presents a significant design compromise for a stealthy platform begins when the length of the fuselage reaches 45 to 50 feet and the empty weight about 30,000 pounds. Above 60 feet, a cockpit essentially presents no impact to optimal aircraft shaping for low-observable properties.
As for cost, the weight of the cockpit would be the largest driver; as mentioned above, CSBA puts it at 4 to 6 percent of the unit cost of an LRS-B-class aircraft. The cost of the extra equipment is small in comparison: All essential systems are likely to cost on the order of $500,000 for cockpit gear and $2 million to $3 million for redundant communications. There is no additional cost for triple- and quadruple-redundant aircraft control and mission management systems, as no modern aircraft would come without them, and platforms built to survive combat are inherently dynamically unstable and require digital fly-by wire and redundant mission management systems.
Such costs come to more than half the $4.5 million list price of a system like the Predator. But for an aircraft with a unit cost of half a billion dollars, the fraction falls to less than 1 percent, so low that aircraft designers might consider such a figure negligible.
Assuming the Air Force plans to buy 80 to 100 LRS-Bs at a target unit cost of $550 million, it falls in this category. If a manned LRS-B is favored, the additional cost to allow it to be remotely piloted would increase the price by less than 1 percent (mostly for communication equipment), representing an opportunity cost of one aircraft (99 instead of the planned 100).
If a purely unmanned LRS-B is favored, the most conservative opportunity cost would be six aircraft (94 versus 100), assuming that cost scaled directly with the additional weight of the cockpit. However, this probably overestimates the case; aircraft designers interviewed for this article said the need for forward ballast would likely cancel out the imagined weight savings of eliminating a cockpit. That suggests that the cost of including a cockpit might actually approximate the cost of the cockpit systems, an opportunity cost well below 1 percent.
There are other ways to look at optional manning’s effect on an aircraft program.
If the aircraft is conceived of as primarily unmanned, designing in the capability for manned operation would greatly simplify the overall test program. Aircraft designers interviewed for this article said the difficulty of simultaneously testing the flight envelope and control software is widely underestimated, and having onboard aircrew might shave 12 months from the time it would take to prove out a purely unmanned aircraft. Estimates vary about just how much money streamlining the development and testing phases might save, but they go as high as $500 million.
Moreover, basic envelope expansion, propulsion and other testing could be conducted without the risk imposed by robotic or tele-operated control, a consideration particularly important when the cost of the test aircraft is high. While some flight-test mishaps are caused by pilot error, humans have proven remarkably capable in novel and unexpected situations and provide significant redundancy and flexibility. If an onboard pilot prevented the crash of a single $550 million aircraft in testing, the inclusion of manned capability would justify itself.
The development-and-testing advantage of optional manning persists long after initial testing. A human on board allows testing of unmanned modes while they are maturing and in airspace where that would otherwise be prohibited. If a new challenge or adversary weapon emerges before there is time or budget to develop autonomous countermeasures, commanders can return to manned operations while unmanned capabilities are expanded.
Finally, the support costs for a purely unmanned aircraft can be substantial. Conversations with the remotely piloted aircraft global communications architects at Air Force Space Command suggest that while existing planned satellite networks could support simple protected command and control, supporting a persistent ISR mission would require a satellite constellation roughly double TSAT in satellite size and number. That would add tens of billions to the $50 billion LRS-B program, or erode the total aircraft buy. Dependence on these supporting programs introduces the vulnerability that if they are cut, the program is no longer viable. Having an initial manned capability at least defers this cost and does not put the capability at risk due to the cancellation of supporting programs.
On the flip side, designing in the ability to gradually fly the aircraft without aircrew aboard will likely reduce manpower and training costs over its decades of service. (Recall that roughly 70 percent of an aircraft’s program cost is not in acquisition but operation; currently, the Air Force spends about 43 percent of its “blue” dollars on operations and support and 23 percent on manpower.) As innovations increase autonomy and allow one operator or crew to control multiple aircraft simultaneously, they promise, long term, to reduce the required number of aircrew and the overall training burden.
Overall, aircraft with missions that drive their range-payload-survivability requirements to aircraft sizes that exceed 30,000 pounds and fore-aft lengths that exceed 50 feet incur only marginal costs by adding optionally manned capability. Most large aircraft acquisition programs, including the LRS-B, meet these criteria. (It is also worth considering that there may be a cost-payback point below this weight; for example, if adding a cockpit shortened the testing period so much that it compensated for the increased aircraft weight, or if the speeded development of unmanned capability saved money in the long term. The actual point at which an optionally manned capability achieves acquisition or life cycle break-even deserves serious study and analysis.) Including remote operation and autonomy requirements in the earliest stages of concept development represents an investment we can expect to pay dividends. Among them, perfection of the optionally manned architecture and the ensuing human-supervised expansion of autonomy gained with the LRS-B will be transferrable to future aircraft systems, both military and commercial. As such, the LRS-B represents a critical transition aircraft leading to more complex unmanned capability in larger, multirole aircraft.
Therefore, the announced decision to make LRS-B optionally manned represents a well-considered decision for the nation, providing it with maximum operational flexibility at acceptable cost, investing wisely in future capability, and taking advantage of a strategic opportunity to significantly expand key unmanned technologies and autonomous capabilities without holding the program, or mission need, at risk to them.