5
The Research Infrastructure at NGA

BACKGROUND

The committee feels that, as with its agency precursors, the National Geospatial-Intelligence Agency’s (NGA’s) ability to meet future mission requirements depends now, and will even more so in the future, on geospatial science and technology research. NGA-led research has been conducted over the years through a wide variety of organizational means. How research is conducted is just as important to its success as what is researched; therefore the committee felt that how research can most effectively be conducted at NGA was one of the “hard problems” to be addressed. This becomes even more important since NGA’s research role is growing to the extent that most major research activity in geographic information science now has some roots in NGA-funded programs, and other federal agencies such as the U.S. Geological Survey (USGS) and the National Science Foundation (NSF) have decreased their allocation of research support to the geospatial disciplines. This chapter reviews the current research framework and makes recommendations for future form, scale, and synergies.

The introductory chapters of this report set out the mission and operational context within which NGA works. These have evolved considerably over the last two decades, with the terrorist attacks of September 11th creating an additional impetus, leading to a paradigm shift in the technology foundations necessary to fulfill NGA’s mission. NGA’s emphasis of operations has moved from imagery collection and map production (i.e., imagery intelligence; mapping, charting, and geodesy)



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency 5 The Research Infrastructure at NGA BACKGROUND The committee feels that, as with its agency precursors, the National Geospatial-Intelligence Agency’s (NGA’s) ability to meet future mission requirements depends now, and will even more so in the future, on geospatial science and technology research. NGA-led research has been conducted over the years through a wide variety of organizational means. How research is conducted is just as important to its success as what is researched; therefore the committee felt that how research can most effectively be conducted at NGA was one of the “hard problems” to be addressed. This becomes even more important since NGA’s research role is growing to the extent that most major research activity in geographic information science now has some roots in NGA-funded programs, and other federal agencies such as the U.S. Geological Survey (USGS) and the National Science Foundation (NSF) have decreased their allocation of research support to the geospatial disciplines. This chapter reviews the current research framework and makes recommendations for future form, scale, and synergies. The introductory chapters of this report set out the mission and operational context within which NGA works. These have evolved considerably over the last two decades, with the terrorist attacks of September 11th creating an additional impetus, leading to a paradigm shift in the technology foundations necessary to fulfill NGA’s mission. NGA’s emphasis of operations has moved from imagery collection and map production (i.e., imagery intelligence; mapping, charting, and geodesy)

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency for pre-definable theaters of war and deployment scenarios to a focus on predictive, on-demand geospatial intelligence (GEOINT) and the ability to respond in near real time, anywhere, anytime. The evolution, however, is resisted by the complexities and immediacy of NGA’s mission imperatives. The emerging GEOINT concepts of “full-spectrum collection,” “horizontal integration,” and “persistent surveillance” (NGA, 2004a) both inspire NGA’s research and development portfolio and simultaneously create a mismatch between NGA’s operational systems and the future requirements for research. The committee observed that there has been a strong tendency to focus research on the improvement of existing architectures. Yet NGA can only achieve so much by investing in research that is based on incremental improvement of data sources and processes within its existing technology. The paradigm shift required to fulfill NGA’s GEOINT mission will unavoidably involve discontinuities in the established technology frameworks, organizational structures, and processes of today. The increased reliance on industry for new technology that has worked well over the last decade may not be sustainable. In industry, this is a time of considerable vulnerability. Many commercial organizations will fail and be replaced by newer organizations unencumbered with past technologies. Yet NGA also has to respond to current tasking. As such, NGA will have to invest in two different streams of research. First, there is a need to maintain and improve existing capabilities to make them as productive and robust as possible. A second need is to confront the realization that meeting NGA’s current, let alone future, mission will require fundamental, as-yet-undefined redevelopment of information technology (IT) infrastructures and operational processes. In the opinion of the committee, the current level of research support is barely sufficient for the first, let alone the second. Nevertheless, NGA has the potential to build on its existing research model to respond to this critical national need, should the nation decide that such a priority is indeed at the heart of the national interest and award support concurrent with that need. It will be essential to allow vastly increased feedback from the existing research process into NGA’s operational programs, and vice versa, so that NGA’s problems can become better known to those conducting the pertinent research. The dichotomous nature of NGA’s research future has provoked a debate about incremental versus fundamental change that is clearly active within NGA, and both strands of the debate were heard during the committee’s briefing sessions. Chapter 2 describes the current structure for geospatial science and technology research at NGA. The present chapter discusses the committee’s findings and conclusions about what is needed to meet the new NGA mission. At the end, this chapter addresses the challenge of moving research forward in parallel with incremental

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency improvement; the migration of “new-generation” research toward an operational solution that will replace the current architecture; and the specific role that could be played by the external research community. RESEARCH ORGANIZATION FOR GEOINT2 The collection of research and development (R&D) support mechanisms used by NGA and described in Chapter 2 has led NGA to the leadership position worldwide in geospatial technologies and capabilities. After reviewing the wide variety of research funding mechanisms, the committee felt that the strongest aspect is NGA’s ability to leverage commercial interests and to collaborate with private-sector organizations and national laboratories to rapidly develop new geospatial technologies. While academic involvement has been strong, the formal NGA Academic Research Program (NARP) component is somewhat dwarfed by the remainder of the structure. Meeting the needs of NGA in the GEOINT2 era will require even closer collaboration among government, industry, and academia in its interactions with the intelligence community (IC). With the changing mission and needs, during the compilation of this report the committee increasingly felt that certain enhancements would help make the research program more effective and responsive to these changing needs. The following sections describe these suggested enhancements. R&D Coordination NGA has now assumed a leadership role in funding geospatial research in the United States. The committee felt that a significant increase in research funding through existing or new programs will be necessary for NGA to build a national research infrastructure and then draw upon this established base to fulfill its vision. NGA clearly appreciates the value of both basic and applied research, as well as the need for a trained and educated future workforce both for its own needs and for those of the nation. Indeed, GEOINT2 and its demands have implications for the whole educational research infrastructure of the United States, from universities and colleges to commercial companies and government agencies. While these entities have a huge stake in the achievement of NGA’s vision, they are nevertheless relatively unguided in how to respond to NGA’s needs for the future. Furthermore, although undirected basic research has high potential payoff, and therefore is beneficial to support, the committee sensed a disconnect between the basic research currently going on in disciplines critical to NGA and what is needed for GEOINT2. Similarly, there seem few means by which NGA’s customers and users can direct their research needs and problems back through the NGA

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency research infrastructure, providing the sort of user feedback that is often necessary for successful business-style operations. No standing or centralized body currently coordinates NGA’s R&D priorities and projects for GEOINT. A coordinated R&D advisory group with stable membership and regular meetings that focuses specifically on defining an overarching architecture, changing priorities, resource constraints, prior R&D investments, and R&D investment plans under way could be advantageous in achieving NGA’s vision. Creation of a new collaborative among the IC, academia, industry, and government around geospatial science would benefit from a “board of directors” with authority to build links and move resources to create synergies. The committee notes that NGA offers a plethora of mechanisms for research support across academia, government, and business; yet these appear to the committee to be uncoordinated and oriented toward shorter-term “incremental” technological approaches to NGA’s challenges. NGA’s InnoVision Directorate could create a permanent coordinating committee with representative internal and external membership, tasked with seeking out, supporting, and coordinating R&D that contributes toward NGA’s vision and needs, at some appropriately low level of national security classification. This coordinating committee could evaluate NGA’s research process itself: comparing strategies, creating triage lists and priorities, selecting topics for each Broad Area Announcement (BAA) or NARP, and debating about those strategies that would best suit different types of research. A coordinating committee could also produce a road map for future strategic research planning. The principal advantages of such a committee would be to (1) increase the proportion of projects that move from basic to applied; (2) raise awareness of the importance of research in achieving NGA’s vision; (3) help build and coordinate the broader collaborative that extends beyond the IC and incorporates its differing needs into research planning; (4) eliminate redundancy by helping to ensure links among groups doing similar research within and outside the government; and (5) help streamline the somewhat disparate research efforts on which NGA depends. On the other hand, such a committee could (1) lead to security problems (few academics, for example, hold the clearances necessary to assist such a committee), (2) run the risk of discouraging “outside-the-box” thinking in research, (3) reduce the effectiveness of the existing NGA chain of command, (4) involve international partners in projects better left in-house, and (5) dilute research efforts being directed from the top as national priorities. While such a standing R&D coordination body could provide some sort of high-level peer oversight, peer review is nevertheless essential to the future quality of NGA research programs, however they are funded. According to a recent Office of Management and Budget (OMB) report,

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency peer review “can increase the quality and credibility of the scientific information generated across the federal government” (OMB, 2004). The committee feels strongly that the peer review process is as essential in NGA geospatial science as in any other science. In its definition of peer review, Wikipedia (http://en.wikipedia.org) notes: “Publications and awards that have not undergone peer review are likely to be regarded with suspicion by scholars and professionals in many fields.” During discussions with investigators of current NARP projects, it was apparent that there was a lack of clear understanding about why their proposals had been selected, or why others had not, and there was little feedback on the research despite site visit contact with program managers. This is particularly problematic since the BAAs and calls for proposals offer few details of why the research is of interest to NGA. The committee felt that NGA is in danger of creating unnecessary and false suspicion around its research due to the lack of peer feedback and review. Despite the perception that lack of peer review is due to security concerns, virtually none of the NGA investigators to whom the committee spoke felt limited in their ability to publish or openly discuss their research as a consequence of receiving funds from NGA. RECOMMENDATION 13: Establish peer review processes whenever possible in order to enhance the effectiveness of the research proposal process. This should include but not be limited to review of solicitations, review of grant proposals, and review of cooperative research and development agreement and partnership deliverables. Lastly, the coordination of NGA R&D programs requires the definition of roles for each type of organization. A clear role needs to be specified for academics vis-à-vis the product vendors and system integrators that play into the overall technology life cycle. While NGA has organizational means in place for funding academic organizations (e.g., NARP), most of these means could be directed to any type of research organization. Yet businesses and academia have different and more subtle roles to play than simply fulfilling the specifications of a contract, not the least of which are educating the next generation of experts and creating market-driven technologies. The division of labor among research efforts and the appropriate relationships among them matter greatly. It is the job of an R&D coordinating function to provide this clarity of roles and responsibilities. Effective coordination could save money, reduce effort, increase the likelihood of success, and reduce the risk of high-risk research projects. The committee was impressed by the annual NARP symposium and found the introductory day of short presentations a worthwhile occasion for others within NGA and its related agencies to learn about NARP-

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency funded research. Nevertheless, the appearance was that little else was done within NARP to ensure that the results of NGA-funded research found their way throughout the agency. There would be significant benefits from closely matching NARP projects and scientists with specific NGA programs, and from promoting and disseminating scientific results within NGA in general. This is a complex task, involving the entire collaborative of groups involved in NGA research. In part, it could involve making reporting and dissemination more central to NARP projects, just as NSF has expanded the role of promoting the broad significance of the research it funds. Consequently, the committee recommends the following. RECOMMENDATION 14: Define clear roles, responsibilities, and relationships for the various types of organizations that conduct and disseminate results from R&D in NGA’s priority areas (e.g., universities, research institutes, national labs, product companies, system integrators, consortia) in order to increase the effectiveness of this multifaceted research program. With respect to Recommendation 14, the committee discussed the NSF’s Integrative Graduate Education and Research Traineeship (IGERT) program and found in it a good model for the simultaneous training and educational advancement of scientists of interest to NGA and for the advancement of NGA’s research goals. One immediately attainable goal for NGA under this recommendation would be to work with NSF, perhaps using a permanent liaison, to promote or fund one or more IGERTs that explicitly meet NGA’s future GEOINT needs. This could be done either by collaborating with NSF or by using the program as a model within the existing research structures. Other federal agencies find themselves in similar situations, and much could be gained by agency collaboration in clarifying these roles. Needs for Development: A Consistent and Flexible Architecture NGA’s mission statement ties success closely to meeting the needs of its users and customers. This group varies from the broader intelligence community to the military services and coalition partners. It now includes agencies more concerned with public safety and homeland security. If NGA’s research is to meet the needs of its future customers, efforts must be directed toward reducing the barriers between research and development. This effort will have the benefit of reducing the time gap from research project to working technology. NGA’s users can help this process by providing feedback, but to succeed, theoretical research must take

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency place in the context of NGA’s customers’ problems, so that the successes of research can quickly transition to systems. One area that the committee felt would help improve this transition is a better understanding by researchers of NGA’s current systems architecture. A computing and information architecture is the suite of hardware and software components for a particular task, plus the human infrastructure and knowledge base needed for their effective use, including the data models implemented and the paradigm, theory, or ontology embodied. Based on interviews with NARP principal investigators and the committee’s knowledge of the NGA research process, it appears that the next-generation target architecture is not clearly defined or promoted by NGA to its R&D community, nor is a baseline architecture defined by NGA for use by its R&D community. Having a better understanding of the overall environment would help researchers and developers more effectively direct and transfer the results of their research to meet NGA’s needs. In lieu of a well-articulated architecture for research purposes, research projects will tend to focus on currently interesting, but isolated, analytical capabilities. The unintended consequence of this is that many of NGA’s R&D investments lead by default to incremental improvements to the existing technology baseline, rather than to the next-generation National System for Geospatial Intelligence (NSG) called for by the NGA vision and doctrine. Also, without a clear understanding of the overall architectural environment, it is difficult to effectively propose research projects that are likely to be transferable to operational systems. Every R&D project is different and can lead to different “artifacts,” including application programming interfaces (APIs), commercial off-the-shelf (COTS) plug-ins, schemas, architectures, system products, papers, graduate student mentoring strategies, and more. Yet, for instance, a researcher might deliver a COTS plug-in to an application within NGA’s technology baseline, but fail to deliver a robust or thread-safe API that might be used in an autonomous process within an enterprise spatial data infrastructure. To integrate the results of related projects under the current architecture, NGA must devote extra resources to defining and communicating its architecture. The need for this expenditure could be obviated by developing exemplar or template architecture guidelines to which external researchers must adhere in delivering project results. The committee notes that relatively few NARP projects transition from three-year research to five-year development activities. NGA can benefit from assisting more projects that lead to tangible NGA-related products or prototypes. In order to get the most benefit from R&D projects in terms of eventual technology transfer, the committee recommends the following.

OCR for page 65
Priorities for Geoint Research at the National Geospatial-Intelligence Agency RECOMMENDATION 15: Define and publicly articulate the current and future geospatial information systems architecture at a level of detail sufficient for researchers to design projects that are easily integrated with it. RECOMMENDATION 16: Be explicit about how the results of R&D projects can be incorporated into current and future architectures, and provide administrative support to researchers and developers to ensure that they are connected with the appropriate NGA staff and contractors to facilitate technology vetting and transfer. International R&D Coalition Although most of NGA’s R&D programs do not forbid the involvement of non-U.S. nationals, companies, and institutions, there is little involvement in NGA’s R&D portfolio by citizens and corporations from coalition countries. This has the consequences of loss of benefit from strands of research being actively addressed by external research groups, under-appreciation of the research directions and potential future architectural developments of allies, and given a shortage of first-rate geospatial research expertise globally, loss of an intellectual contribution from leading research groups that could accelerate NGA’s research time scales. The committee noted several instances in which minor problems resulting directly from the ability or inability to share information and techniques became sufficient to restrict the scope or execution of research. The committee therefore recommends the following. RECOMMENDATION 17: Work to broadly involve the geospatial science and technology R&D community from coalition countries, to ensure that NGA has access to the broadest possible pool of expertise.