Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 79
Basic Research in Information Science and Technology for Air Force Needs 7 Priorities in Basic IS&T Research for the Air Force It is the committee’s consensus that the basic research topics described in Chapters 3 through 6 are of great importance for enabling the capabilities sought by the Air Force and that none of them are likely to be adequately explored by the private sector. Recognizing, too, that AFOSR’s resources for IS&T are relatively limited, the committee recommends in this chapter how to prioritize among those research topics given different funding scenarios. As background for that discussion, it first presents its overall vision for Air Force basic research. A MODEL FOR AIR FORCE IS&T BASIC RESEARCH Given the limited availability of 6.1 funds for IS&T research, the committee believes it is best to give priority to those research areas that AFOSR gives a realistic chance of enabling significant new Air Force capabilities. This is not to say that all of AFOSR’s basic research in IS&T should be tied to particular technology goals—in fact, such tie-ins should be an unusual exception—but that IS&T basic research should normally be motivated by the recognition of a gap in the knowledge base that underpins desired Air Force technologies. The committee suggests that these knowledge gaps be seen as “grand challenges,” which is a helpful way to motivate research and build research communities and a useful mechanism for communicating the relationship between basic research and future Air Force technologies. Topics designated as grand challenges would normally be ones that are properly addressed by a cross-disciplinary community of basic researchers, and such a designation gives that community an identity and
OCR for page 80
Basic Research in Information Science and Technology for Air Force Needs strengthens its coherence. Grand challenges should be defined in terms that are recognizable to the basic research community, but AFOSR should also be able to map the grand challenges to future Air Force technologies, strengthening the linkage between basic research and the rest of Air Force R&D. The grand challenges are not part of, nor do they compete with, the AFRL focused long-term challenges (which are more oriented toward technologies), but they should link to them. Some examples of grand challenges in a general sense are included in a recent paper by well-known computer scientist Butler Lampson:1 Development of a computer that hears, speaks, and sees as well as a person; A system architecture that scales up by 106; and An information system that can be used by millions yet only requires a part-time support staff. Such grand challenges quite naturally facilitate new interdisciplinary research communities: “interdisciplinary” because the breadth of the challenges calls for varied expertise and “naturally” because the associated researchers are interested in the whole range of efforts addressing the grand challenges. The Computing Research Association has for several years held conferences that similarly seek to identify grand challenges for computer science more generally. Information about these meetings may be found at http://www.cra.org/grand.challenges/. Figure 7-1 suggests how the grand challenges model would contribute to cohesion across the total Air Force R&D enterprise. The grand challenges would be defined by AFOSR as a clear response to AFRL and Air Force technology needs, but in terms of concepts that motivate the basic research community. They have the advantage of not biasing the research directions chosen to address a particular Air Force need: They define a challenge for the research community that relates to desired capabilities but do not present the research community with specific technology goals, as would be the case if the basic research program were motivated directly by current Air Force technology programs. In turn, the 6.1 research builds up a base of knowledge related to the grand challenges (the arrow on the lower right of Figure 7-1), and the rest of the Air Force R&D community can track progress on the grand challenges that support its technology goals. That transition (indicated by the arrow at the lower left of Figure 7-1) is aided by the computational laboratories recommended in Chapter 9. 1 B. Lampson, “Computing meets the physical world,” The Bridge 33(1) (2003).
OCR for page 81
Basic Research in Information Science and Technology for Air Force Needs FIGURE 7-1 Mapping of Air Force needs to basic research through grand challenges.
OCR for page 82
Basic Research in Information Science and Technology for Air Force Needs The characteristics that are important in defining grand challenges for the AFOSR are these: They should be representative of important systems that the Air Force would like to have but that are well beyond the state of the art, like the examples in the next paragraph. The key component technologies that would make them realizable should be technologies where the Air Force wants research advances. They should be readily comprehensible to researchers who do not have domain-specific knowledge of the Air Force system. They should bound the research somewhat, so that researchers can aim toward specific applications rather than general capabilities. The committee recommends that AFOSR consider the following as possible grand challenges, but the list is by no means exhaustive: Control of multiple UAVs. Research to enable the control of multiple UAVs by one human in mixed manned and unmanned airspace, in contrast to today’s requirement for many humans for a single UAV in carefully deconflicted manned and unmanned airspace. Taskable airborne network. Research to enable cost-effective and rapidly deployable tactical intelligence networks in urban environments, where the nodes generally are sensors carried on UAVs or lighter-than-air vehicles and the networks are taskable by ground-and air-based commanders. Mixed-reality training environments. Research to enable training for air crews, command post staff, and commanders in an environment of such fidelity that it would be indistinguishable from the real world (and in fact would sometimes involve the real world—hence “mixed” rather than “virtual” or “augmented”). The computer tools used in such training environments should be the same as those used in the real world. An automated Air Operation Center staff assistant. Research to enable software that can learn from being told, much as human staff members learn on the job. Rapid system integration. Research to enable the rapid integration of IT-based systems, such as those belonging to different members of ad hoc coalitions. This research would encompass HSI, networks and communications, security, software, and information management. It is envisioned that an interdisciplinary community would engage in collaborative research focused on each grand challenge. Researchers would describe repeatable experiments and metrics with which to assess
OCR for page 83
Basic Research in Information Science and Technology for Air Force Needs the research results. Greater synergy with 6.2-6.3a research programs could result in shared research environments or computational laboratories. Funding for 6.1 research should continue to be used to find solutions but not (normally) to develop or demonstrate particular technologies. RECOMMENDED BASIC RESEARCH PRIORITIES IN IS&T The committee recommends that in determining program priorities, AFOSR should emphasize research that clarifies Air Force problems, impacts operational procedures, and advances the state of the art in areas of interest to the Air Force. Prior collaboration with Air Force laboratories should not be essential to obtaining program support. AFOSR should also continue to monitor the emergence of possibly disruptive technologies or Air Force missions and adapt the research program and strategy accordingly: The rise of network-enabled systems and the emergence of a global war on terrorism have been recent disruptive changes, and they will not be the last. The reader is reminded that the committee did not examine the entire portfolio of the AFOSR Mathematics and Space Sciences Directorate, which manages most of AFOSR’s IS&T research, and it did not mean to imply that other topics covered by that directorate are less important. Rather, the topics funded by the programs in physical mathematics and applied analysis, computational mathematics, electromagnetics, and space sciences were not within the scope of this study, and it is not the committee’s intent that IS&T topics should be emphasized at the expense of those topics in other programs. Even greater care is necessary in interpreting the priorities given here that overlap the current AFOSR programs in dynamics and control, optimization and discrete mathematics, high-performance computing and advanced architectures, and signals communication and surveillance, which overlap with IS&T only in part. Thus, any high-priority research identified by the committee in Chapters 3 through 6 that would be funded by one of those programs was not weighed against other topics covered by the same program, only against other IS&T topics. For instance, challenges in the control of complex ad hoc networks, distributed information systems, and multiple UAVs were considered by the committee in setting priorities, but it did not weigh in the importance of other challenges in control, such as control theory for avionics. Recommendation. The committee consensus on how to prioritize Air Force IS&T research in networks, communications, information management, software, and HSI under different funding scenarios is encapsu-
OCR for page 84
Basic Research in Information Science and Technology for Air Force Needs TABLE 7-1 Relative Priorities Under Four Funding Scenarios IS&T Topic 10% reduction Stable 10% increase 25% increase Networks H H H H Communications H H H H Information management M M H H Software L M M H Human-system interactions H H H H NOTE: “H” means the general topic is a high priority, and its funding should be protected or increased. “M” means the general topic is of medium priority for AFOSR support, given the contributions by other players, but not of medium importance to the Air Force. “L” means funding in that area should be sacrificed so that a critical level of effort can be supported in other areas. “L” does not mean that the topic is not of importance to the Air Force, only that if resources are tight, it is a reasonable candidate for cuts because other organizations are contributing to the area and/or the challenge is so great that a small AFOSR effort is unlikely to lead to significant progress. These priorities pertain to the five general research areas listed in the left-hand column as weighed only against one another, not against other programs funded by AFOSR’s Mathematics and Space Sciences Directorate. The priorities are meant to show the committee’s consensus on which of the areas to (de)emphasize with any change in funding for their support. They take into account not only the importance of the research but also the relative need for Air Force-specific research. They reflect the committee’s general sense of what can be meaningfully accomplished within the funding scenarios posited. The committee did not develop a detailed estimate of the resources required for each of the research topics in the left-hand column. lated in Table 7-1. See Chapters 3 through 6 for detailed recommendations about research directions in each of these five areas. With the current funding available for IS&T (the column labeled “Stable”), the committee recommends that networks, communications, and HSI portfolios merit the highest priority, while the information management and software research portfolios would be better able to weather any forced reductions in effort. The committee is not saying that the latter two research areas are less important to the Air Force. Rather, it is the committee’s judgment that if cutbacks are required, reductions in those programs would do the least harm in limiting future options. If the overall IS&T funding dropped by 10 percent, the committee would give software the lowest priority only because other organizations, and commercial enterprises, are performing some related research. If overall funding increased by 10 percent, the priority for information management research should be increased. Finally, if overall IS&T funding were to increase by 25 per cent, the committee recommends a balanced portfolio drawn from the recommendations of Chapters 3 through 6.
OCR for page 85
Basic Research in Information Science and Technology for Air Force Needs Because all five major research areas listed in Table 7-1 contribute synergistically to the future fielding of team-focused, network-enabled systems, progress toward that vision is dependent on a balanced research effort across all five areas. However, overall funding would have to increase significantly to support such an effort, a conviction that is contained in the following recommendation: Recommendation. The committee recommends a significant increase in IS&T funding within AFOSR, centered around research to support team-focused, network-enabled systems of interest to the Air Force.
Representative terms from entire chapter: