Accessibility Information

Users of assistive technologies such as screen readers should use the following link to activate Accessibility Mode before continuing: Learn more and Activate accessibility mode.

A-SpaceX Industry Day Announcement

Solicitation Number: A-SpaceX
Agency: Department of the Air Force
Office: Air Force Materiel Command
Location: AFRL/RIK - Rome
  • Print

Note:

There have been modifications to this notice. To view the most recent modification/amendment, click here
:
A-SpaceX
:
Modification/Amendment
:
Added: Aug 14, 2008 1:35 pm
The purpose of this updated announcement is to: 1) Invite potential offerors to engage in private discussions with the Analysis WorkSpace for Exploitation (A-SpaceX) program management as follow-on to the Industry day that was held July 8, 2008 ; and 2) Make available electronic copies of the materials from the Industry day. In order to receive the most current information relating to A-SpaceX we ask that all potential performers be registered with A-SpaceX management team by sending an e-mail with your contact information and affiliation to: DNI-IARPA-ASPACEX@UGOV.GOV. All A-SpaceX Industry attendees are already registered. Parties seeking one-on-one communications should send an e-mail to the A-SpaceX management team at: DNI-IARPA-ASPACEX@UGOV.GOV . When appropriate, please indicate if the subject of the communications relates to contractual questions or technical questions by including the keyword: CONTRACT or TECHNICAL in the email subject line.



Background.



The Air Force Research Laboratory (AFRL), on behalf of the Office of the Director of National Intelligence (ODNI) and the Intelligence Advanced Research Projects Activity (IARPA) hosted an Industry Day July 8, 2008 prior to soliciting proposals for innovative, creative, and high-risk research to advance the state-of-the-art in technologies and methods for the Analysis WorkSpace for Exploitation (A-SpaceX) program. The anticipated BAA will address Phase 1 of what is being conceived as a multi-phase program. A BAA is expected to be released following the Industry Day, and will provide additional details regarding the research opportunities within this program. The BAA will solicit proposals for conducting critical research and development for what is planned as the first phase of a multi-phase, multi-million dollar program. Phase 1 will have an expected performance period of 18 months, and would culminate with a technology showcase for the benefit of potential users and system integrators. Succeeding phases would focus on developing integrated systems. Formal proposals will be reviewed by an IARPA selection committee based on source selection criteria to be published as part of the BAA. Separate proposals may be submitted for different requirements with the thrust areas, e.g., Time Machine and MindSnaps. We anticipate initial contract awards will be made by the first quarter of 2009. AFRL is expected to be the Contracting Authority for the planned work, while IARPA will be responsible for technical oversight and the BAA solicitation process.



Overview:



A-SpaceX is seeking to create an analytic environment where the workspace becomes an enabler for the analytic process – fostering creativity and effective analytic reasoning. A-SpaceX envisions two key emerging technologies being fused for the intelligence analyst workstation of the future: Virtual worlds, and Workflow management.

Analysts will work with an integrated workstation that not only allows access to the legacy tools of today, but any number of virtual worlds designed to support the analytic process. Virtual worlds are defined here as: 1) Complex multi-dimensional digital environments (i.e. models) that are shared across many users; 2) They may incorporate data models from a variety of sources, made up of objective data as well as hypothetical and/or theoretical constructs; and 3) Both persist and evolve whether or not any individual user is actively engaged with that world. A-SpaceX will demonstrate that this technology affords transformational opportunities to change the way people work in general, and how analysts will work in particular. Synthetic worlds will be multi-dimensional, dynamic visualizations of data tailored to support different analytic problems. A-SpaceX anticipates focusing of the development of two Synthetic worlds that will: 1) Highlight the dynamic aspects of both factual and synthetic (hypothetical) data and models and the benefits of combining them in an integrated representation that provides a meaningful context; 2) Allows the changes in the data to be systematically explored so as to facilitate both forensic and proactive analysis; 3) Show the dynamic aspects of cognition, specifically the process of thinking about data as reflected by capturing decision states that can serve as mental bookmarks for the analytic process; and 4) Demonstrate how groups of analysts can collaborate in the exploration of data and the sharing of analytic strategies.



The A-SpaceX vision allows for multiple synthetic worlds, each of which is optimized to support different kinds of analytic decision-making. Multiple synthetic worlds will be optimized for different analytic problems. The incorporation of multiple, coherent virtual worlds will require that there be a consistent set of protocols and usability standards to ensure that analysts may readily jump from one world to the next as needed to support exploring complex, inter-related data.



Implicit in the A-SpaceX vision is premise that changes in the data used in defining a synthetic world will be available to automated agents working on behalf of the analyst to both find relevant information, and highlight when the data (already integrated into analytic products derived from that synthetic world) have changed. Accommodation for legacy software applications and analytic tools will be integrated into the workspace through the incorporation of “proximal displays” located on desktops close to the analysts, while exploration in parallel synthetic worlds will be available through “distal displays” on the periphery of the analyst’s workspace will serve as windows into the synthetic world. Though used for different purposes, the proximal and distal displays will be part of a single integrated digital desktop so that no matter how physical displays are used, the moving of information from one display area to another is as seamless as possible.



Phase 1 of the A-SpaceX program as currently conceived will have two major thrusts: 1) The research and development of key science and technologies to create a “Time Machine” synthetic world that allows complex data to be explored in a rich context, as well as allowing changes in data to be explored over time. 2) The research and development of science and technologies for tracking an analysts workflow to derive “MindSnaps”, or mental bookmarks that reflect key aspects of an analyst’s decision-making process in conducting analyses, as well as the development of a Decision Space virtual world designed to enable and analyst’s visualizing his or her own analytic process via MindSnaps. Time Machine and MindSnaps development is dependent upon a Synthetic World foundation which includes an over-arching concept of operations for the employment of virtual worlds as part of the analytic infrastructure, along with the definition of prospective architectures for the use of synthetic worlds with legacy software and hardware infrastructure.



Time Machine Thrust.



Virtual worlds are in fact models that have several relevant properties: they can be shared by multiple people, they are persistent, and they evolve over time. We anticipate that a virtual world may be built based on data derived from the physical world, as well as including models, events and other factual data. In addition, the data may be updated as a model, it is possible to systematically explore temporal aspects of change of data and events that occur in the physical world. Forensic analysis can be performed by replicating the real world in a synthetic world. In this forensic mode, alternative explanations can be explored by adjusting assumptions and the quality index of the collected data and allowing the changes to propagate into the future. Proactive analysis could be explored by applying predictive models that look forward in time and suggest indicators leading to future events.

Virtual worlds are interesting as they allow the representation of complex, n-dimensional data sets; they can be shared by multiple people, and they evolve over time. In developing synthetic worlds to facilitate analysis, one of the more interesting aspects of synthetic worlds will be their ability to explore how data changes, and therefore the implications of that change can be used in formulating hypotheses about, and explanations of, change. We believe a key dimension of exploring changing data will be the ability to manipulate time in the synthetic worlds – in effect turning these worlds into Time Machines. The use of synthetic worlds as Time Machines is conceptually about the explorations of data, and as such, may be (but need not be) tied to a physical location. These synthetic worlds will allow a variety of data abstractions to be embedded, layered, and compared. A critical issue in creating a Time Machine capability will be the ability to detect and manage significant changes from data. A-SpaceX will seek research and development for technologies that will address contextually relevant change detections and management from raw data streams. Further, the complexity of the data sets possible in synthetic worlds for analysis suggest that tools to optimize how those worlds work and appear for specific analytic problems will require the development of context “lens & filters” that will make it possible to highlight, distort or block different kinds of data within a synthetic world.



The CONOP for Time Machines is expected to transform how analysts would explore data. For instance, analysts would begin their analytic work by selecting a “here and now”— where they pick a perspective of interest from which to explore the data sets represented within a synthetic world. The analysts would then assess possible futures relative to their chosen perspective in place and time. This would become the basis for effectively assessing the possibilities for a “then and there” relative to the “here and now” and developing potential hypotheses for what is possible in the future given what is known and unknown from the “here and now”. Another analytic CONOP might be that analysts would hypothesize a discrete (specific) future and then assess the viability of that hypothesis by exploring both known and unknown data represented in the synthetic world. In effect, the Time Machine CONOP would allow analysts to engage in forensic and proactive analysis, by viewing past/future perspectives.



MindSnaps Thrust.



Current information technology is completely agnostic regarding helping an analyst to perform work. The burden is on the analyst to use tools to annotate, organize, and associate their work products. The analyst’s workflow process is complicated by methodology practices and security requirements of the IC that add even more tedious and time consuming activities. Today, there is no formal or proscribed mechanism for tracking analytic workflow. Current automation limitations make it extremely difficult for analysts to remember, find, and return to their analytic decision points to reconsider information changes, or try alternative hypotheses. None the less, analysts have to explain, share and propagate work products with supporting rationale to peers and managers.



The analytic process is inherently iterative and fractionated. The analyst will typically think

about one aspect of analytic problem until it is no longer fruitful, new data comes in, or a new facet of the problem becomes more salient. Analytic decision making is iterative because the analyst will often return to a previous point in his or her thinking and resume analysis from that point in time until a solution is found to the analytic problem. We believe that there is a theoretical and technological basis to monitor an analyst’s workflow, and detect when the analyst has shifted their “decision trajectory” while engaged in this iterative, fractionated process. Further, when a shift in decision trajectory is detected, the system would collect meta-data regarding the analyst’s workspace including what software applications were in use, where they were located and the data sets open in those applications. By examining the work that was being done before the decision trajectory shifted, it should be possible to extract meta-knowledge about what the analyst was doing during that period of analysis. The meta-knowledge would become the basis for creating a meaningful abstraction to summarize that decision trajectory. Together, the meta-data about the workspace and the meta-knowledge data would be combined to create a MindSnap. MindSnaps will be the basis for the creation of mental bookmarks of an analyst’s decision making. A-SpaceX seeks to develop a research thrust that explores creating a synthetic world that is based on MindSnaps and facilitates an analysts exploring the analytic process tied to a problem of interest. This "Decision Space" synthetic world effectively maps the overall decision-making processes of analysts as graphical decision nodes, and allows them to find and, cognitively re-load their workstation from different points in their own analytic process.

Common to both the development of Time Machine and MindSnap capabilities will be the development of an overarching concept of operations (CONOP) and a shared Human–Computer Interface for the use of synthetic worlds as analytic tools. Further, the application of current virtual world technologies to support the proposed CONOP will require the adaptation of current virtual world technologies. The A-SpaceX program will likely, therefore, include a systems engineering task to address the common information technologies anticipated as being required for the use of synthetic worlds in support of the analytic tasks. During Phase 1 of the A-SpaceX program, we anticipate that these tasks will be largely limited to the definition of coherent CONOP as well as architecting the requirements for supporting systems architectures. There may be limited development of those services and the definition of knowledge representation in support of the critical services identified as being required for the A-SpaceX CONOP.



Summary.



An A-SpaceX Phase 1 Industry Day was held on July 8, 2008 . The purpose of Industry Day was to provide the A-SpaceX vision and concepts / ideas that may appear in the BAA, and solicit feedback from potential performers on key ideas and technical challenges. Parties still interested in participating in the A-SpaceX program or commenting on the program described above are welcome to communicate with the A-SpaceX team prior to the publishing of a BAA by sending an e-mail to: DNI-IARPA-ASPACEX@UGOV.GOV. You will then be contacted to arrange a brief conference call with the A-SpaceX management team. Please be advised, the management team cannot discuss or evaluate specific research or proprietary technologies. Conversations relating to the A-SpaceX vision, general research A-SpaceX interests, or potential technical / theoretical challenges and significant development milestones are encouraged.



We cannot control the types of problems that future analysts might face. We cannot control the demands for understanding information and the pressure for faster decision making. We can, however, provide the analyst with an environment that encourages creativity, reasoning, collaboration with internal and external experts, rapid exploration of alternate perspectives for greater understanding, and detecting change in data and through processes within a multi-dimensional information rich synthetic world. Welcome to A-SpaceX where it’s about… the future.



THIS ANNOUNCEMENT is for information and planning purposes only and is not to be construed as a commitment by the Government. This is not a solicitation announcement for proposals and NO contract will be awarded from this announcement. No reimbursement will be made for any costs associated with providing information in response to this announcement or any follow-up information requests.



POC & Contact information for A-SpaceX:

Dr. Jeffrey Morrison, PhD is the Program Manager and the IARPA point of contact. dni-iarpa-aspacex@ugov.gov



Mr. Peter Rocci of AFRL is the principle POC for the AFRL contracting office. peter.rocci@rl.af.mil.



Contracting Office Address:

AFRL/Information Directorate 26 Electronic Parkway

Rome, New York 13441-4514



Please consult the list of document viewers if you cannot open a file.

Industry Day Briefing and Questions and Answers

Type:
Other (Draft RFPs/RFIs, Responses to Questions, etc..)
Label:
Industry Day Briefing and Questions and Answers
Posted Date:
August 14, 2008
aspacxbriefing.pdf (3,530.83 Kb)
Description: Industry Day Briefing, 8 July 2008
ASpaceXQA.pdf (25.67 Kb)
Description: Summary of Questions and Answers, 15 July 2008
:
AFRL/Information Directorate
26 Electronic Parkway
Rome, New York 13441-4514
:
Peter Rocci