The UCAV Ascendancy: What are the Problem Issues?

This paper explores a number of fundamental technology issues which will present obstacles to Uninhabited Combat Aerial Vehicles fulfilling a broader portion of the mission spectrum currently performed by manned combat aircraft. These obstacles are then discussed in the context of Australia’s AIR 6000 program. In the context of Australia’s public debate on program AIR 60002, the UCAV frequently assumes mythological proportions. Indeed, a former Minister of Defence advocated UCAVs as a possible alternative to manned aircraft. There can be no doubt that much of the enthusiasm for UCAVs as manned fighter replacements in Australia stems from budgetary pressures to which the RAAF, and the ADF as a whole, are being subjected. The prospect of ‘robot fighters’ which do not incur the large cost overheads of fast jet pilot training and retention, or the difficulties in recruiting suitable candidates for pilot training, is clearly attractive to many participants in this debate. The essential argument for the UCAV is centred on a perceived lower cost of ownership and fleet operation, in theory allowing the RAAF to field a substantial combat force without the budgetary grief traditionally associated with manned fighters. With the RAAF facing very significant funding challenges in fielding the AIR 5077 Wedgetail ‘Pocket AWACS’, the AIR 5402 tanker replacement, and ultimately the AIR 6000 F/A-18A replacement, any alternative which is perceived to yield a major reduction in ‘bucks per achievable bang’ will inevitably develop a following. This paper will explore some of the fundamental technological issues which present obstacles to UCAVs assuming roles currently performed by manned fighters. Roles and Missions The essential starting point for any comparative discussion of the merits of manned aircraft against UCAVs, or vice versa, must be roles and missions to be performed. A necessary and essential condition for the replacement of any capability with an 1 Paper prepared for the `UAV Australia’ conference, 8&9 February, 2001, Melbourne, Australia. 2 AIR 6000 aims to replace the F/A-18A Hornet and possibly the F-111 Aardvark. Refer Figure 1. alternative is that the substitute can effectively do the same job, and ideally do it either better for the same expense, or cheaper. Figure.1 AIR 6000 aims to replace the F/A-18A Hornet, and possibly the F/RF111C/G tactical fighters (Author). If we consider the role spectrum performed at this time by the F/A-18A Hornet in Australian service, we will find that the aircraft is primarily used for counter-air operations, with secondary strike roles in which it supplements the larger F-111C/G. These roles can be further subdivided thus: • Offensive Counter-Air flying Combat Air Patrols into contested airspace to destroy opposing fighter aircraft, and supporting assets. • Defensive Counter-Air flying Combat Air Patrols over the air-sea gap or Australian territory to destroy opposing aircraft and cruise missiles. • Maritime Strike supplementing the F/RF-111C/G and P-3C as Harpoon ASCM delivery platforms, and providing fighter escort for the former in contested airspace. • Close Air Support / Battlefield Air Interdiction supplementing the F/RF111C/G in striking at opposing ground forces either in contact with or in the proximity of Australian ground forces, and providing fighter escort for the former in contested airspace. • Defence Suppression supplementing the F/RF-111C/G in striking at opposing air defences, and providing fighter escort for the former in contested airspace. • Interdiction and Strike supplementing the F/RF-111C/G in striking at opposing strategic assets and road/rail communications, and providing fighter escort for the former in contested airspace. In essence, the F/A-18A will fight for control of the air, and once this is achieved, swings to supplementing the F/RF-111C/G in attacks on surface targets. The role spectrum for the F/RF-111C/G is no less challenging in its breadth and its depth: • Offensive Counter-Air Strike destroying hostile aircraft on the ground, air bases and other supporting infrastructure. • Defence Suppression destroying hostile air defence assets, especially radars, C3 and SAM/AAA systems. • Strategic Strike destroying hostile strategic military assets, such as ballistic missiles, C3, port facilities, fuel supplies etc. • Maritime Strike destroying opposing maritime surface assets using the Harpoon, AGM-142E and guided bombs. • Close Air Support / Battlefield Air Interdiction striking at opposing ground forces either in contact with or in the proximity of Australian ground forces. • Interdiction striking at opposing road/rail communications. • Defensive Counter-Air interception of opposing maritime patrol aircraft and transports at large operating radii. Both the F/A-18A and the F/RF-111C/G will rely heavily upon the Wedgetail AEW&C aircraft and aerial refuelling tankers. The capability goals in the December 2000 White Paper are quite specific, in that they define the geographical footprint for such operations to encompass the air-sea gap to Australia’s north and north west, and for strike operations any base or asset within the region which could be used to threaten Australia or ADF forces operating in the region. A UCAV replacement for either or both the F/A-18A and F/RF-111C/G would therefore have to cover an exceptionally broad role spectrum, and do so with no loss in flexibility, over a geographical footprint which necessitates aerial refuelling on almost every single mission profile. This is no mean feat by any measure. Figure 2. Cyberdyne T-800 Terminator, the central protagonist of James Cameron’s `Terminator’ sci-fi thrillers, is perhaps the best contemporary popular representation of the self-aware autonomous robot warrior ( Photo copyright Universal). Wherein do the Obstacles Lie? There are no fundamental technological obstacles in aerodynamics, structures, propulsion, flight controls and low observables technology which would preclude the design of a UCAV capable of matching or exceeding the aerodynamic performance, load carrying capability and combat radius of either the F/A-18A or the F/RF111C/G. Given the dictates of Breguet’s equation, such a UCAV would be similar in size and weight to the aircraft it is replacing, given that the weight and volume of the pilot and mission avionics are not substantial in relation to the gross weight of combat aircraft in this category. To deliver an F-111 sized payload to the radius of an F-111 will require a roughly F-111-sized UCAV, at a similar cost in airframe technology. The much bigger issue to be resolved here is that of replacing the decision-making entity, which is the pilot or pilot and navigator, and the mission avionics used to support the flight crew. Current UCAV literature identifies two idealised extremities in implementation. One is a ‘dumb RPV’ wholly controlled in every respect by a remote human operator over a datalink. The alternative is a wholly autonomous ‘robot fighter’ which in the manner of James Cameron’s ‘Terminator’ is assigned a target to kill, and does so with the cunning of a human pilot, and the error free precision hoped for of true artificial intelligence (AI). Refer Figure 2. To wholly replace a modern manned multirole fighter across its role spectrum, a UCAV must in the limit conform to either of these hypothetical paradigms3. Figure.3 With increasing autonomy the demands for datalink capacity and thus bandwidth to a UCAV will decrease. In the limit, a UCAV with human-like cognitive and reasoning ability would require no more datalink bandwidth than a manned combat aircraft (Author). The ‘dumb RPV’ model essentially amounts to using a digital pipe to transfer what the pilot of a manned aircraft sees from his cockpit over hundreds of nautical miles or more into the remote cockpit of an operator in a controlling aircraft, or ground installation. It must do so without loss of information, thus imposing challenging demands in bandwidth, and also do so with exceptional reliability in a hostile jamming environment, thus further exacerbating bandwidth needs. To establish a reasonable bound on required data channel capacity, a throughput of the order of a high definition television (HDTV) picture would arguably be reasonable. Using lossy compression techniques, this is typically of the order of tens of Megabits per second, per UCAV. Using lossless techniques for mission critical imagery and data, the capacity demand could be much higher. If we assume an aggressive jamming environment, a spreading ratio of the order of 1000 or more for a noiselike modulation may be required. This imposes bandwidth requirements of the order of 10 GHz per UCAV. Such bandwidth requirements are arguably not 3 The alternative could be argued to be some balance between the non-autonomous and fully autonomous extremes. The difficulty in such a hybrid is where to set the `datum point’ for splitting the decision-making intelligence between the UCAV and the remote controller. implementable over satellite microwave links, given the established antenna and transmission technology base, and the need for both redundancy and the concurrent support of multiple UCAVs in a given area of operations. Figure 4. Moore’s Law for microprocessor transistor counts, on a logarithmic scale. The exponential growth rate in component complexity is plotted here for a range of commercial processor chips. Transistor counts around 1 billion can be expected around 2005 (Author). Tropospheric propagation constraints are not conducive to the use of millimetric band channels of such bandwidth over the distances associated even with Low Earth Orbit satellite channels. Propagation delays of milliseconds preclude genuine real-time control loops, and thus operator inputs must amount to commands to an autopilot in the UCAV. In the limit, the ‘dumb RPV’ would not appear to be economically implementable if it were to wholly replace a modern