| 2006 Technology
Symposium >Artificial Intelligence
Artificial Intelligence
The Artificial Intelligence technical area investigates technologies,
tools, and processes that support the discovery, processing, exploitation,
and dissemination of information, tools, and knowledge. Intelligent agents
are covered in this area.
A Multi-Source Recommender System for Intelligence Analysis
Tom Bartee, Principal Investigator
Location(s): Washington and Bedford
Problem
Intelligence analysts have an enormous amount of information available to support their investigations. Unfortunately, analyst-driven information searches may miss important information. Searches are time-consuming, and the information and supporting applications change constantly. Furthermore, analysts are prone to human judgment biases that can lead them to miss or discount information that does not directly support their hypotheses.
Objectives
We will develop a Recommender System that supports intelligence analysis through the automatic retrieval of information that is relevant to the specific current needs of the analyst. This information will save analysts time and facilitate unbiased decision making by providing a range of data that would both support and refute analysts' working hypotheses.
Activities
We will implement the core infrastructure for instrumentation of analysts' activities, identification of key events, and browsing of information retrieved by the system. Next we will develop an Expert Task Model for our core challenge problem and implement Event Processor scripts to automate searches. Finally, we will move our system into an analytic environment and perform a formative evaluation.
Impact
Our approach for a Task-Oriented Recommender System will be applicable to a broad range of intelligence problems, as well as to problems outside the intelligence community. The generic architecture supports software reuse across domains.
^TOP
Audio Hot Spotting for Tactical and Intelligence Applications
Qian Hu, Principal Investigator
Location(s): Washington and Bedford
Problem
Increased capability to capture audio and video sources requires automatic systems to rapidly filter and prioritize audio information. Depending on the embedded speech recognition technology, current COTS systems either miss critical information or present excessive false alarms, clogging the analysis queue. Furthermore, COTS systems cannot search on critical non-lexical audio cues. Phoneme-based COTS systems cannot search across languages, requiring linguists for all query creation.
Objectives
Our objective is to build on audio hot spotting (AHS) technology developed on a previous project to automatically identify segments of interest from multimedia sources within and across selected target languages.
Activities
We plan to (1) combine multiple automatic speech recognition technologies to improve audio retrieval accuracy and recall; (2) adapt and adjust our non-lexical cue detection algorithms; (3) extend the cross-lingual search capability to Arabic; and (4) improve efficiency by automating the end-to-end AHS system, including batch query processing and user alerts for easy operation and deployment.
Impact
The proposed research and development will improve the accuracy and efficiency of AHS, allowing users to receive reliable time-sensitive indications and warnings more quickly. The knowledge and experience gained will serve to (1) lead/advise industry in developing AHS capabilities tailored to intelligence applications, and (2) guide and evaluate AHS systems and embedded technologies.
^TOP
Cooperative Robot Pairing
Bob Grabowski, Principal Investigator
Location(s): Washington and Bedford
Problem
The military services and the DoD are counting heavily on future robotics capabilities. The current paradigm focuses on the development and execution of a single robot platform type. Mission effectiveness is constrained by the size, processing, and sensing of that platform. These limitations may unnecessarily limit the range of missions unmanned systems can support.
Objectives
This program proposes to enhance existing robot functionality by combining robots across different scales for a more robust mission capability while simultaneously providing low-level autonomy to reduce the dependency and duty cycle of human operators. Specifically, we plan to coordinate two of our existing robot platforms, the Meteor and the PackBot, to execute disposal of an improvised explosive device.
Activities
We will conduct several field exercises to verify our capabilities. Possible strategies include coordinating with the Future Combat Systems program to get possible participation in Experiment 1.1 or 1.2, participating in the NATO Military Operations on Urbanized Terrain exercises scheduled for June 06, and working with the Marines on experiments to measure our team's performance against that of traditional approaches.
Impact
We will demonstrate the utility of cooperating robot platforms that reduce the demands of a single platform and increase the utility of unmanned systems in mission scenarios. We will develop individual robotic functionalities that can transition rapidly to the military community. This will allow us to provide actual field platforms to test and refine the future of unmanned system warfare.
^TOP
Coordinated Financial Investigation
Conrad Chang, Principal Investigator
Location(s): Washington
Problem
"Sharing is about networking people, building relationships, and fusing information," according to the Markle Task Force. Many vital investigations, such as criminal money laundering, would be more effective if we could connect people working related cases and topics. This capability is urgently needed to move from reactive investigation of crimes to coordinated financial investigation about illicit activities.
Objectives
Our goal is to create a match-making capability for investigators with financial crimes, based on their current investigative needs. This capability will proactively suggest connections that might benefit each participant. We will prototype how such a capability can cross both system and organizational boundaries and can respect individual and organizational information sharing policies.
Activities
We will begin with generic infrastructure to enable services and virtual data to span communities, and then develop match discovery algorithms and processes to identify and inform people working on similar cases. We will prototype a policy mechanism for each individual and organization to specify criteria for sharing and evaluate the effectiveness of our methods using customer-provided and open source data.
Impact
This research can immediately benefit multiple agencies involved in financial investigation by automating much of the human connection process. It can also benefit the IRS in its efforts to track down companies that are evading taxes.
^TOP
Inference Rules for Joint Mission Assessment
Lewis Loren, Principal Investigator
Location(s): Washington and Bedford
Problem
Planning and execution disconnections across strike and ISR missions detrimentally impact the Air Force's responsiveness. Currently, there is no efficient means of determining ATO mission status or an aircraft's munitions load and, while the target list is constantly evolving, such changes frequently fail to find their way to the ISR deck. Such shortfalls prevent a real-time assessment of combat objectives.
Objectives
Prototypes will be developed to address planning and execution shortfalls by making inferences regarding mission status and combat objectives. Such inference rules will also improve situation awareness for operators controlling ISR assets, strike assets, and logistics support, and will support effects-based operations by enabling a mapping between real-time activities and the objectives and timetable.
Activities
We will develop prototypes with interfaces to mission planning systems. These prototypes will make inferences regarding mission status and combat objectives. A beta version of the prototypes will be ready by the second quarter, and will be presented to operators for feedback and assessment by the third quarter.
Impact
The developed prototypes will permit MITRE and ESC to respond quickly to pervasive problems of great importance to operators. We will demonstrate the value of inference rules and their applicability to time-sensitive targeting, ISR management, and effects-based operations. Stakeholders include MIT, the C2 Battlelab, and mission planning systems.
^TOP
ISR Forensics
Curtis Brown, Principal Investigator
Location(s): Washington and Bedford
Problem
Forensic analysis will become one of the future drivers of both exploitation tool development and multi-intelligence (multi-INT) data archives. To be successful in exploiting the ISR data in large multi-INT databases, forensics analysts will need to combine ground-moving target indicator data with data from other sources of intelligence and will require easy-to-use tools to access, navigate, and process it.
Objectives
Our objective is to create visualization tools to support the forensic analysis of ISR data. These tools will provide a framework for exploring automation through the incorporation of advanced tracking technology.
Activities
We will develop a set of tools that exploit geospatial and temporal data, and then apply the tools to synthetic ground truth data sets that we will generate. Advanced tracking technology, including Multiple Hypothesis Tracker traceback, will be developed and integrated to aid the analyst. We will document lessons learned and demonstrate the forensic environment at the Technology Symposium.
Impact
We will make our exploitation tools available to analysts. Lessons learned will be fed back both to multi-INT database efforts as well as exploitation tool development efforts.
^TOP
Persistent Surveillance for Border Security
Russell Graves, Principal Investigator
Location(s): Washington and Bedford
Problem
Persistent surveillance must be pursued in the context of an application: we must use exactly the right mix of sensors, with the right modalities and the right sampling rates to characterize the activity of an adversary. We bound our problem by focusing on the infrastructure and operations of the smuggling rings conducting illegal business near the United States-Mexico border.
Objectives
Our objectives are to understand smuggling operations and track their activity through integration of intelligence and operational information with multiple sensor modalities tailored to the time stability of the events in the smuggling chain. The ultimate goal is to provide decision makers with actionable and timely information to help them develop their interdiction and surveillance strategies.
Activities
We will apply forensics analysis to GMTI data from the Tethered Aerostat Radar System using tools including MITRE's ISR Forensics and the Analysis and Feasibility Tool. We will seek opportunities to integrate the tools and the analysis methodologies into agencies' operational centers, and explore adding other sensors and integrating/fusing intelligence data and suspicious activity reports where possible to help achieve persistent surveillance.
Impact
The development of a persistent surveillance methodology through integration of operations and intelligence in the context of the border security problem could have a direct impact upon the human and narcotic smuggling rings. What we learn from the United States-Mexico border region can be generalized to apply to other national and worldwide border security issues.
^TOP
Robot Swarms
Bob Grabowski, Principal Investigator
Location(s): Washington and Bedford
Problem
Most unmanned systems require teleoperation by dedicated human supervisors. Tasks such as reconnaissance call for teams of robots and make a one-to-one paradigm inappropriate. Challenges include ways to manage and integrate different types of robots, task and coordinate despite incomplete knowledge, execute a collective mission with a distributed set of teams, and provide appropriate control and feedback to the human operator.
Objectives
Robots must increase their current level of autonomy while more effectively leveraging the collective information of the team. This project will investigate the fusion of local autonomy coupled with globally derived information to execute a specified mission. Ultimately this research will reduce the degree of human intervention required to execute and monitor a given mission.
Activities
We will develop an integrated simulation engine for testing large robot teams, but most effort will be directed toward real robots working in real indoor environments. We are developing a building takedown scenario that employs multiple, heterogeneous robots to detect intruders, deploy sentries, and secure a building space. Robots must work closely together to overcome individual limitations and maintain connectivity in the space.
Impact
In the future, our sponsors (military, intelligence gathering, homeland security) will need to deploy multiple robots of differing types and capabilities, but ensure they function together as a single logical unit. The research in this project will provide strategies that allow multiple robots to work together more effectively.
^TOP
Structured ISR Fusion
Walter Kuklinski, Principal Investigator
Location(s): Washington and Bedford
Problem
The importance of developing fusion algorithms to improve the warfighting capability derived from Intelligence Surveillance and Reconnaissance (ISR) systems is well documented. Many operational requirements mandate levels of performance that can only be obtained by fusing data from multiple sensors and platforms. However, only limited attempts have been made to develop formal design procedures for data fusion systems.
Objectives
The objective of this MSR is to use formal mathematical methods such as Random Set Theory and Category Theory as frameworks to design and implement fusion algorithms and to predict fusion system performance. We will develop and implement a widely applicable methodology for fusion system design and analysis.
Activities
We will develop and evaluate, using both simulated and field data, multi-target tracking algorithms that fuse conventional radar SMTI data with additional data sources including SIGINT, COMINT and ELINT, using both Random Set Theory and Category Theory frameworks.
Impact
The successful completion of this MSR will both provide a significant advance in the general area of fusion system analysis and design, and yield specific fusion algorithms applicable to the future fusion requirements of many MITRE sponsors.
^TOP
|
