The relative impact on today's National Airspace System by its five major user segments—airlines, air cargo, general aviation, business, and military--is shown in the accompanying pie chart as percentages of IFR air traffic volume handled by the ARTCCs. This accounts for virtually all airline, cargo, military, and business flying, but less than half of general aviation, with its high proportion of VFR flying. [Because each of these segments is changing at a different rate, similar pie charts were also created for 2025 and 2050, based on forecasts made from the data in the table immediately below them.]
  

The degree to which unmanned aviation will impact NAS operations over the next 25 to 50 years depends on the timing and level of its acceptance in each of these segments, assuming these segments' shares remain relatively unchanged. It is apparent from this diagram that a large acceptance of unmanned aircraft in the near term by the military or business aviation segments would still not significantly affect NAS operations, whereas limited acceptance of them, even in the far term, in the airline segment would produce a significant impact. The perspective embodied in this last statement is critical to objectively assessing the potential of unmanned aviation to impact the future infrastructure of the NAS because the perspective (and expertise) of today's unmanned aircraft community (and the FAA) is focused solely on only one of these segments, the military. The following discussion attempts to quantify the extent to which unmanned aviation will likely impact the NAS in each of these five segments over the coming quarter to half century.

Airline Traffic. The major challenges to airlines accepting unmanned aircraft (as in passenger flight without a pilot onboard) are technology availability, pilot unions acceptance, and customer acceptance. The major technical hurdles are automating see and avoid, approach and landing, and ground operations.
     Although an automated see and avoid system capable of avoiding midairs with cooperative as well as non-cooperative traffic is not currently commercially available, components of such a system are undergoing flight evaluation today (see the 2006 OSD/USN See & Avoid Flight Demonstration and the 2006 USAF/AFRL See And Avoid Flight Test experiments) and should be certified and available in a commercial package in the $20,000-price range within the next 5 years. TCAS provides such a capability today, but only for cooperative traffic and is not certified to tie into any flight control systems; its installed price is in the $150,000 to $250,000 range. Only two airline midairs (Germany, 2002 and Brazil, 2006) have occurred since the introduction of TCAS following the 1986 midair over Los Angeles, and it is noteworthy that one accident involved a pilot acting contrary to his TCAS advisory and the other to a pilot not having his TCAS turned on.
     When the Los Angeles ARTCC lost its Voice Switching and Control System for 3 hours on 14 Sep 2004, ten of the 800 airborne flights in its sector violated FAA separation standards in the first 15 minutes, underscoring the single point failure implications of ground communications in today's NAS. Three points emerge from this incident, 1) human and TCAS see and avoid capabilities, even together, are still insufficient without ground (ATC) cueing to ensure safety, 2) Los Angeles ARTCC would have remained able to communicate with pilots of unmanned aircraft when it was not possible to do so with airborne pilots in this situation, and 3) this would have been a non-incident in the Free Flight environment being developed by FAA. TCAS is an interim measure, and the human eye is inadequate in certain collision scenarios, especially as air congestion increases. Automated, non-cooperative see and avoid is critical to manned aviation safety; it is not solely a requirement for unmanned aviation.
     Of all phases of flight, landing is the most demanding of piloting skills and, by extension, the most challenging to automate in unmanned aircraft. Category 3 ILS-coupled auto-land, a ground-centric capability at selected major airport runways, has been available to most airliners for nearly two decades. More recently, Boeing's and Rockwell Collin's Global Landing System (GLS) has demonstrated an auto-land capability, using GPS combined with ground-based Local Area Augmerntation System (LAAS) functionality, for any runway at any airport worldwide. Tests showed the ability to consistently auto-land within 3 feet of runway centerline, even in equatorial regions where GPS signals are least consistent. Both FAA and JAA have certified this system. Commercial availability of the Rockwell Collins GLU-925 GPS/LAAS receiver is planned shortly. As it must also be with future see and avoid systems, manned aviation safety was the dominant driver for GLS achieving FAA certification
     Of the three technology areas identified, ground operations has had the least effort undertaken on behalf of unmanned aircraft. However, its implementation can draw on the previous two technologies discussed, see and avoid for ground obstacle avoidance and landing for precision taxiing. This capability should be certified and commercially available in the 2010-2015 timeframe, assuming manned airliner ground operations are the dominant push.
     Pilot unions accepting the removal of their members from airliner cockpits will arguably be the strongest challenge faced by unmanned aviation being accepted in the passenger-carrying role. History is against them, however. The chart below illustrates the decline in cockpit crew size since the introduction of passenger service. This trend has occurred due to technical advances (INSes eliminated navigators, etc.) and its acceptance eased by the thin (or negative) profit margins and frequent pilot layoffs inherent in the airline industry. The trend predicts zero cockpit crew sizes in the post 2020 timeframe. While that may be realistic for air cargo, various factors will likely push this date out by a decade or two for airliners, but "roboliners" will debut sometime in the 2025-2050 period. Before the two-pilot crew gives way to the one-pilot one, pilot unions will undoubtedly raise the spectre of the single pilot being incapacitated inflight by food poisoning or heart attack. Unfortunately, pilot suicides (SilkAir, 1997; EgyptAir, 1999) have proven to be a more frequent occurence than medical anomalies. Robots are not subject in either case.

     The recent experience of the railroad industry in adopting remote control locomotives (RCLs) for use in railyards is instructive on this point. After a decade of fierce union opposition, some 10 percent of U.S. locomotives are RCLs. In that time, the accident rate in railyards (which itself accounts for half of all U.S. rail accidents) has been cut in half, meaning total rail accidents have been reduced by 25 percent. Railyard-incurred injuries have been reduced by 57 percent, and the average cost per railyard accident by 34 percent. Union opposition, originally focused on accident rates increasing, has now shifted to expressing concern about the ergonomics of the RC-equipment harness and electromagnetic radiation hazards from the RC transmitters (less than from cell phones). For those who associate technology leadership with America, European railyards operate three times as many RCLs, and Canada began using RCLs a decade before the U.S.

     Public acceptance of roboliner flight is inevitable and will not be the major barrier envisioned by many in the unmanned community. When elevators were introduced in department stores, elevator operators were employed to reassure the shopping public of their safety until the reliability of elevators was established in their mind. Has the reliability of elevators changed significantly between then and now? No, but the public's acceptance of them has. Similarly, the flying public has already accepted the elimination of registered nurses as flight attendants and the inclusion of women as pilots. When both pilots of a recent USAirways flight reported for duty intoxicated, their flight's passengers still filed onboard without questioning the sobriety of their replacements. When boarding, most passengers never even look left into the cockpit to see if the pilots are there. Those same passengers will unhesitatingly board a driverless tram to take them to other terminals. The fact that roboliners will not be subject to highjacker or terrorist demands may even make them preferable in some travelers' eyes. The flying public trusts the FAA to ensure the airlines are safe, so the FAA, not the public, may prove to be the "acceptance" barrier, if any. Once unmanned cargo aircraft demonstrate a record of safe, reliable flight over, perhaps, a decade, roboliner acceptance will follow.
     In summary, the airline segment is driving the few remaining enabling technologies needed by unmanned aviation to enter into any segment; the last of these technical barriers (ground operations) will be overcome by 2010-2015. Union and public acceptance will follow in 2025-2050 once the FAA allows it. The airline segment presents the superset of barriers to unmanned acceptance, but each barrier will have been overcome one at a time in other segments by the time roboliners appear. Once the technical hurdles are overcome (2015), the FAA will dictate when, and industry economics how quickly, this transition occurs.

2025/2050 Forecast: Of the airline segment's 55 percent of total IFR air traffic handled in the NAS, 0 percent will be unmanned, or 0 percent of the total volume. By 2050, roboliners will have been introduced and will constitute between 0 and 50 percent (assume 25 percent) of airline IFR operations, or 14 percent of total IFR traffic.

Air Cargo Traffic. Well recognized within unmanned circles as its "great commercial hope," air cargo freighters seem to possess all the attributes--long, dull, overwater missions--and none of the drawbacks --no passengers, hubs at secondary (Class C) airports--desired to initiate unmanned operations. They present only two of the three major challenges inherent in the airline segment, technology availability and union acceptance, and the latter as a much lower hurdle. The technology challenges are the same as those postulated for the airlines, and indeed the cargo segment is an equal driver for the auto-land capability. The cargo segment will likely begin transitioning to single pilot operations in the 2010 decade and unmanned operations in the 2020 decade, initially on long haul, transoceanic routes, followed by the longer overland routes, then the remainder. American (NAS) unmanned cargo operations will probably follow their introduction on Asian routes by 5-10 years. Asia is a likely proving ground because approaches to its major airports are overwater (Singapore, Hong Kong, Narita, Inchon) and the air cargo market is experiencing its greatest growth in that regional market.

2030/2050 Forecast: Of the air cargo segment's 15 percent of total IFR air traffic handled in the NAS, 33 percent will likely be unmanned, or 5 percent of the total volume. By 2050, it should constitute 100 percent, or 15 percent of the total.

General Aviation Traffic. General aviation pilots fly for the personal satisfaction of flying, i.e., there is no motivation in this segment currently to transition to unmanned operations. Indeed, the present position of AOPA (as an organization, not necessarily its individual members) toward unmanned aviation is that it will pose a threat to the safety of its pilots. It fears the current generation of UAVs (non see and avoid equipped), if allowed routine access to the NAS, raising the potential for midairs with its members as much as the superior see and avoid capabilities of future UAVs raising the standard for them and requiring the cost of adding such a system to their aircraft. The intersection of genav and unmanned interests therefore occurs in the potential for midairs, making the major challenge in this segment the specific, previously addressed one of see and avoid technology. Ironically, the highest midair rate of any segment occurs among genav pilots, making them the primary beneficiary of the see and avoid technology being developed for unmanned aircraft. Making the cost of such a system low enough is key to acceptance by the genav population.
     There is however a potential future development in genav that would drive a significant portion of it to embrace unmanned operations. NASA's Personal Aircraft System (PAS) program envisions a robust market for airborne commuters, i.e., non-pilots and families being flown short to medium distances in smart airplanes, within a generation. PAS is exploring the hardware and software requirements for such aircraft now and finding few technical hurdles enroute. Transitioning harried commuters from their one-dimensional worlds to one of three dimensions may seem quite a leap, but the subject of "mini-jets" presently seems to have equal or greater credibility than do unmanned aircraft in FAA's envisioned future. PAS, or mini-jets, are actually unmanned (or more properly, "fully automated"), in that there is to be no pilot onboard, with the rare exception of those passengers who happen to also be pilots. If PAS/mini-jets do come to fruition, they will help pave public acceptance of roboliners. If a commuter trusts his life daily to an automated aircraft, why not occasionally to an automated airliner? Before dismissing the possibility of a PAS/mini-jet society occuring in the next 25 years as illogical, consider how the logical requirements for a commuter vehicle (single passenger, high gas mileage, low weight for low taxes, small size for ease of parking) have somehow resulted in the large proportion of SUVs during rush hour.

2030/2050 Forecast: Of the general aviation segment's 12 percent of total IFR air traffic handled in the NAS, 0 percent will be unmanned, or 0 percent of the total volume. By 2050, assuming the introduction of a PAS-like market by 2030 and that it grows over 20 years at the same 8.28 percent exhibitied for IFR handling by genav in the mid 1990s, 33 percent of genav, or 4 percent of the total, could be unmanned genav.

Business Traffic. Corporate jets represent aviation's express lane, flying the few at unscheduled times to unscheduled places to save them time. Their passengers were typically driven to and from the airport and expect to continue to be driven between airports, so the business aviation segment will be the last bastion of pilot chauffers. This segment's major hurdle to unmanned operations will be solely customer acceptance. Corporate pilots will be retained as status symbols as much as for any feelings of insecurity with automation. Seemingly, the frequent occurence of high profile accidents in this segment (Golfer Payne Stewart, 1999; Governor Mel Carnahan, 2000; Senator Paul Wellstone, 2002) would encourage an eager and early transition to automated flight.

2030/2050 Forecast: Of the business aviation segment's 8 percent of total IFR air traffic handled in the NAS, 0 percent will be unmanned, or 0 percent of the total volume. By 2050, unmanned business aviatuon could mirror the overall general aviation trend and be 33 percent unmanned, or 2 percent of the IFR total.

Military Traffic. The military segment's adoption of unmanned aircraft faces one major hurdle, technology availability, specifically for the three capabilities (see and avoid, landing, and ground operations) described for the airline segment. Some would add customer (or union-equivalent) acceptance, pointing to what is termed the "white scarf mafia" or mentality, a reference to the objections of some military pilots who see flying with or against an unmanned aircraft as somehow unheroic and demeaning to their profession. Its better connotation is that it represents the conservative voice of progress within the military aviation community, those saying be wary of jumping to new technologies and concepts until they have proven themselves in combat. The pace of unmanned aircraft adoption by the military services reflects this cautious progress. The table below illustrates how the numbers of types and roles of unmanned aircraft have gradually increased with their use in each conflict of the past 15 years. In that time, unmanned aircraft have progressed from a few small types flying only imagery reconnaissance missions to over 20 types of large to miniature unmanned aircraft flying strike, diversion, and base security missions in addition to various modes of reconnaissance.

Today, the U.S. military operates some 200 unmanned aircraft potentially capable of IFR operations and an equal number of smaller, model airplane size (VFR-only) ones. The former represent just over one percent of all military aircraft. By 2010, the former number will triple and the latter will increase eight to ten fold. By 2020, these numbers could increase by similar multipliers again as the Army's Future Combat System and the Navy's Littoral Combat Ship come on line, both planned to be heavy employers of unmanned aircraft. The Coast Guard's Deepwater program envisions 76 of its planned 238 aircraft, or 32 percent, being unmanned by 2016; that number is zero today. These and various other programs' budgets provide a clear projection for the numbers of unmanned aircraft expected to be in the military segment by certain dates, as shown below.

[When compared with the Services' projections of the number of manned aircraft expected to be in their inventories by these same dates, the percentage of unmanned aircraft in this segment can be estimated, as is done in the table below.]

2030/2050 Forecast: Of the military segment's 10 percent of total IFR air traffic handled in the NAS, 34 percent will be unmanned, or just over 3 percent of the total volume, assuming the 2020 projected inventory continues to triple each decade and triples again by 2030 (200 UAVs today times 3 in 2010 x 3 in 2020 x 3 in 2030 = 5400 of the military's 16,000 air and rotorcraft). These percentages will likely continue to increase, to some 50 percent, or five percent of total IFR traffic by 2050.

Summary Forcast

In summary, 8 percent of the IFR traffic handled by FAA in 2030 is expected to be unmanned and, assuming PAS-like inroads into general aviation, 40 percent by 2050.

     Manned and unmanned aviation tend to regard themselves as two separate (some would say incompatible), non-overlapping communities within aviation. The relationship between them should be viewed from the following perspective to gain a better perspective on how the unmanned side will impact the NAS in the future, because in the near future the distinction between the two will become blurred as people become passengers on automated (i.e., no onboard pilot) aircraft..
     Aviation history has exhibited a long, continuing trend of pushing technology to automate the piloting function since its start in 1903. Only 8 years after the Wright Brothers flew (1911), Sperry demonstrated a prototypical autopilot, whose further development (by Sperry's company and others) produced the first commercial autopilots by 1930, 19 years later. These autopilots served as relief pilots, a desireable adjunct to the human pilot onboard but not a flight critical one; pilots began the transition from continuously flying to occasionally being flown. By the late 1970s, the autopilots in the F-16 and Space Shuttle had evolved into fly-by-wire (digital) flight control systems, necessitated by the numerous, continuous control inputs beyond the abilities of their human pilots. They were now flight critical components, having evolved from the first FBW system in the Mercury spacecraft nearly 20 years earlier. Without them, these aircraft were unflyable, so pilots now transitioned from flying to being flown. FBW systems moved into airline cockpits when the Airbus A320 entered service in 1988 without a moment's hesitation from the passenger public. The next step in this evolution occurred in 1986, when DARPA's Condor made the first fully automated flight from takeoff to touchdown, a capability now productionized in Global Hawk (17 years later). Manual piloting in any phase of flight had now become unnecessary. The final step in this evolution will occur when automated aircraft are trusted to carry man as a passenger. The final enabling technologies, autoland and see and avoid, are becoming commercially available in the next few years, again after some 20 years of development. Whether as a stunt or as a serious demonstration, that milestone will almost certainly occur within the next 10 years.