Current intelligence fusion systems are not accurately and quickly performing the intelligence, surveillance, and reconnaissance (ISR) fusion necessary for tracking moving targets that use camouflage, concealment, and deception to avoid detection. Combatant commanders require a more flexible and responsive capability to engage fleeting and mobile targets.
AFRL scientists established the Targets Under Trees (TUT) program to find, fix, track, target, engage, and assess stationary and moving surface targets.1 Now, they are developing the component-level, TUT Intelligence Fusion System (IFS), which provides the capability to fuse intelligence data from multiple ISR sensors. The IFS architecture integrates the capabilities of multiple fusion components and systems to produce actionable information about the location and identity of time-critical, mobile, ground vehicle targets.2
As depicted in the figure, the IFS currently processes six intelligence data types: ground moving target indicator (GMTI) data, electronic intelligence (ELINT), foliage penetration (FOPEN) change detections, human intelligence (HUMINT) free-text reports, unattended ground sensors, and Predator video. Presented with information streams solely from these sources, the analyst must synthesize the amassed data, looking at each report's details and converting the individual reports into target tracks. In contrast, the IFS automatically associates the multiple reports and converts them into fused target tracks.
The IFS relies on a number of mathematical and physical-movement algorithms to produce tracks and classify targets. By thoroughly and rapidly processing the available data, the system provides the analyst a jumpstart on situation awareness. Researchers designed the IFS to correlate tracks from multiple sources and then determine whether those tracks indicate one or more targets. The IFS can also accept raw sensor measurements and either produce a track or associate the measurements with existing tracks. Finally, the IFS posts track data to a database for ready access by other systems and analysts.
Acknowledging the unique value of an analyst's perception and threat understanding, scientists designed the IFS to provide a number of opportunities for the analyst to interact with the system. For example, the analyst can set the GMTI tracker to minimize the automatic fusion of target reports, thereby producing smaller track segments that require analyst intervention to merge into a single track. The analyst can also interact with the system through a number of commands that allow, for instance, merging of specified tracks or updating of the target battle damage assessment. The IFS provides an additional level of flexibility by allowing analysts to create scenarios and limit scenario targets to those that would be expected during a specific mission.
AFRL included the IFS in its demonstration of predictive battlespace awareness and effects-based operation for the Joint Expeditionary Force Experiment (JEFX) conducted in 2004. The IFS was the first system at any JEFX to require realtime access to a large number of intelligence data sources. For the experiment, scientists configured the IFS to parse datasets from the Automated Assistance With Intelligence Preparation of the Battlespace (A2IPB) system into visually meaningful objects for display on the AFRL-developed Web-Based Timeline Analysis System (WebTAS). WebTAS provided analysts substantial flexibility in determining what and how intelligence information was presented. The analyst could elect to overlay the fused tracks on the GMTI; view only the raw GMTI reports; select from a number of preformed queries for the display (e.g., "show the last known fused track position" or "show fused track with history"); specify that WebTAS update the displays at a specific rate for each query; or manually refresh the display or any of the display components. The analyst could also replay track information on the WebTAS display.
While the IFS has the capability to exploit multiple data sources and produce improved target assessments, the availability and tasking of the sensor systems will limit these opportunities. Additional sensor capabilities and enhanced exploitation of the currently available sensor data will improve the ability of the IFS to automatically generate timely reports with increased value to the analyst. The IFS represents a substantial improvement in helping intelligence analysts perform a very difficult job.
Ms. Sharon Walter and Mr. Robert Macior, of the Air Force Research Laboratory's Information Directorate, wrote this article. For more information, visit http://www.afrl.af.mil/techconn_index.asp . Reference document IF-H-06-09.
1 Brown, W., Kaehr, R., and Chelette, T. "Finding and Tracking Targets." AFRL Technology Horizons®, vol 5, no 1 (Feb 04): 9.
2 Drago, D. "Intelligence Fusion System (IFS) for Targets Under Trees (TUT)." AFRL Technical Report AFRL-IF-RS-TR-2005- 379 (Nov 05).