Insight

Today's military intelligence analysts are faced with the monumental and escalating task of handling massive volumes of complex data from multiple intelligence (multi‐INT) sources and types.  All of this data must be evaluated, correlated and ultimately used to support a commander's time-critical decisions and actions.

To enhance analysts’ ability to more effectively and efficiently process information, DARPA’s Insight program seeks to develop an adaptable, integrated human-machine Exploitation and Resource Management (E&RM) System—the next-generation intelligence, surveillance and reconnaissance (ISR) system.  Through the development of semi- and fully automated technologies, the Insight E&RM System aims to provide the following real-time or near-real-time capabilities in direct support of tactical users on the battlefield:

• combination, analysis and exploitation of data and information across multiple sources, including imaging sensors, non-imaging sensors and other sources;
• efficient management of sensor tasking; and
• detection and identification of threats through the use of behavioral discovery and prediction algorithms.

Insight’s E&RM System is comprised of four main components:

• Unified All‐source ISR Human‐Machine Interface to enable analysts to visualize information from multiple perspectives in such a way that it is readily discoverable, accessible, meaningful and useful;
• Adaptive Multi‐source Exploitation System to enrich and leverage available multi-INT data and provide exploitation products to analysts;
• Collection and Resource Management to provide analysts with a common operating picture of ongoing collections, generates service requests, and plans and allocates sensor taskings; and
• Integration, Processing and Data Archive Environment (InsightOS) to provide a central data repository for all E&RM System components; manage and monitor the mapping of dynamic algorithms on the execution environment; and capture the pedigree of all products to support the reasoning behind algorithmic exploitation results.

For developmental purposes, the Insight E&RM System is supported by several test beds, including a:

• Physical Test Bed to exercise E&RM System capabilities with real‐time air and ground‐based sensors and intelligence collection activities in the context of realistic military scenarios set against the backdrop of an operationally relevant environment;
• Virtual Environment Test Bed which provides the ability to scale to realistic real-world complexity, time and space (which is not feasible to demonstrate in the physical test bed); and
• Development Incubator Test Bed which serves as the repository for collected, simulated and processed data.

Insight’s vision for the E&RM System is to:

• replace existing stovepipes with an integrated system that operates across national, theater and lower‐level tactical intelligence systems;
• optimize data handling to effectively leverage existing and nascent ISR technologies;
• be mission and sensor agnostic and have applicability across dynamically changing theaters of operation;
• be standards‐based to make it easy to add, remove, substitute and modify hardware and software components as they are developed and become available for integration; and
• promote efficient collaboration between intelligence analysts and increase the efficiency and effectiveness of individual analysts through a unified global ISR picture.

The Insight program executed a field test in February 2012 which demonstrated end-to-end system functionality and the capability of the E&RM System to perform sequence-neutral (i.e., out-of-order) fusion of data from multi-INT sources, as well as graph-based multi-INT fusion.  The field test also produced a unique, multi-modality, high-fidelity truthed data set which is available to ISR researchers across the Department of Defense and Intelligence Community.  This data set, combined with the foundational data set collected in the Fall of 2010, provides an unparalleled 135-terabyte resource to more than 240 users across government, industry and academia.