This document presents the Department of Defense’s (DoD) roadmap for developing and employing unmanned aerial vehicles (uavs) over the next 25 years

6.1 Operational Metrics

To relate the priorities expressed by the CINCs in section 3 to the technologies coming available within the next 25 years (section 4), a number of operational metrics (see Table 6.1-1) were devised for this Roadmap. To use the road atlas analogy again, the CINCs’ desired capabilities represent destinations, the technologies possible routes to them, and these metrics mileposts to indicate progress toward the requirements. They identify specific opportunities for future capabilities to satisfy the warfighters’ needs. All references to years are for dates when these capabilities are expected to become available for fielding. Some of the capabilities described have already been demonstrated in labs; others, primarily in the communications and processing areas, will soon be emerging in commercial applications.

Availability

2. Achieve 30% increased time-on-station with same fuel load 2010

3. Achieve 40% increased time-on-station with same fuel load 2015

Signature 4. Field a UAV inaudible from 500 to 1000 ft slant range 2004

Payloads Resolution 5. Distinguish armed from unarmed individuals from 4 nm 2002

6. Distinguish facial features (identify individuals) from 4 nm 2005

7. Achieve 12 in SAR resolution over a 10 nm wide swath 2000

8. Achieve 6 in SAR resolution over a 10 nm wide swath 2002

9. Achieve 3 in SAR resolution over a 10 nm wide swath 2005

10. Achieve 3 in SAR resolution over a 20 nm wide swath 2010

Data Links Data Rate 11. Relay entire COMINT spectrum in real time 2005

12. Relay entire ELINT spectrum in real time 2025+

13. Relay 10-band multi-spectral imagery in real time 2000

14. Relay 100-band hyper-spectral imagery in real time 2010

15. Relay 1000-band ultra-spectral imagery in real time 2025+

Information Processor 16. Map surf zone sea mines in near real time 2002

Processing Speed 17. Map surf zone sea mines in real time 2016

18. Reduce DTED level 5 data in near real time 2009

19. Reduce DTED level 5 data in real time 2022
Metrics for endurance and signature reduction were defined to show how future UAV platform performance could be enhanced to meet the CINCs priorities. The endurance metrics seek to provide 20-, 30-, and 40-percent increases in flight endurance by equivalent improvements in specific fuel consumption (SFC) for a given engine type and constant fuel weight. Figure 4.1.2-1 predicts these increases should be attainable, due to such efforts as the AFRL’s Versatile, Affordable, Advanced Turbine Engine (VAATE) program, by 2005, 2010, and 2015 respectively. These percentages equate to 20-, 30-, and 40-percent more time on station for the same number of deployed aircraft used today, helping address the coverage shortfall identified by the majority of the CINCs. The signature metric, driven by the CINCs’ priorities for enhancing force protection, is to provide a UAV that is inaudible from 1000 ft, and ideally from 500 ft, slant range to preclude detection by base intruders. Figure 4.1.2-2 anticipates the mass specific power of fuel cell-powered engines will equal or exceed that of noisy internal combustion engines by 2004, enabling their use in fielding a silent airborne sentry.

To illustrate future payload opportunities, resolution metrics for EO/IR sensors and SARs were developed. Based on a recurring scenario in many theaters—an embassy or non-combatant evacuation from a foreign city—CINCs need a standoff sensor to avoid both further inciting the local populace and/or being downed by MANPADs (e.g., SA-7/14) with maximum ranges of up to 4 nm. Such sensors should be capable of distinguishing armed from unarmed persons, and, ideally, identifying specific individuals. The former capability requires video imagery with a GRD of 4-8 in, NIIRS 8, and an instantaneous field of view of 0.014 mrad; the latter requires a 2-4 in GRD, NIIRS 9, and a 0.007 mrad IFOV. Figure 4.2.2-1 predicts that improved focal plane arrays could enable today’s gimbaled EO/IR sensor turrets to reach these levels of resolution by 2002 and 2005, respectively. For area searches, today’s best SARs can image the equivalent of a 10 nm wide swath at 12 in resolution. The metrics chosen are to halve this resolution (6 in), then halve it again (3 in) for the same swath width, then to double the swath width covered to 20 nm and again achieve 6 and 3 in resolution. Figure 4.2.2-2 forecasts these capabilities being fielded by 2001 (6 in), and 2005 (3 in) for 10 nm wide swaths, and by 2010 (3 in) for 20 nm wide swaths.

Advances in UAV data links were measured in terms of data rate-based metrics needed for relaying unprocessed SIGINT and uncompressed multi-spectral imagery in real time. Such capabilities would contribute strongly to ensuring CINCs receive ISR information inside their opponents’ decision cycles. For SIGINT, the capability to relay the entire COMINT spectrum or the entire ELINT spectrum was chosen. Figure 4.3-2 forecasts the communications technology for these opportunities could be fielded by 2005 and 2025+, respectively. For IMINT, the ability to relay successive 10-band multi-spectral (MSI) at 0.16 Gb/image, 100-band hyper-spectral (HSI) at 1.6 Gb, and 1000-band ultra-spectral imagery (USI) at 16 Gb, all at 1 sec intervals was chosen. These levels should be reached by 2000, 2010, and 2025+, respectively. Of course any decision to increase reliance on lasercoms would potentially allow the necessary data rates to be achieved sooner.

Finally, metrics were developed for information processing based on CINC prioritization of, and emerging technology in, counter mine warfare. The technology is that of ONR’s Airborne Remote Optical Spotlight System (AROSS), which currently employs 500 MHz processing over 48 hrs to extract images of broaching sea mines from Predator UAV Skyball video. After optimizing AROSS’ software, this process will still require an hour between imaging and results being available for dissemination. Using faster processors in the future, this time could be reduced to near real time (20 min) using 1.5 GHz processing, or to real time (1 sec) with 1800 GHz processing speeds. Moore’s Law, as shown in Figure 4.4-1, predicts such levels of processing speed should become available in 2002 and 2016, respectively, but the latter date is within the period during which the limits of silicon-based processing will reach its limits, so real time mapping of sea mines may have to depend on alternative forms of computing. The same limit is encountered to reduce IFSAR data to DTED level 5 maps of a 150 x 150 nm area in near real time (2009) and real time (2022). Both examples illustrate an important caveat of the trends developed in section 4 and applied to operational tasks here, that of recognizing the limits to a given technology’s growth.

The upper half of Figure 6.2-1 plots the predicted appearance of these 19 metrics over the next 25 years, with the date of each centered within a 5-year window of estimated initial availability for fielding.