16 AI MAGAZINE
IF node or the aggregate processing of a suite of
IF nodes, each with similar structure. Figure 1 is
strictly a discussion aid and not an architecture or
The JDL model (Hall and Llinas 1997) and subsequent revisions (Steinberg, Bowman, and White 1999;
Llinas et al. 2004; Blasch et al. 2006), were proposed
to provide a useful categorization representing logically different types of problems, generally (although
not necessarily) solved by different techniques, and
to maintain a degree of consistency with the mainstream of technical usage. Much of the discussion on
IF models is drawn, either directly or in modified form,
from the revisions to the JDL model by Steinberg,
Bowman, and White (1999). We will use the following definitions:
Level 0 — Subobject data assessment. Estimation
and prediction of signal/object observable states
on the basis of pixel/signal level data association and characterization. This is a new level
added to the JDL process model, ascribed to the
data sources. One clear example of this type of
processing is the detection of blobs of interest
in imagery, which signify an unknown entity of
interest not having a semantic label.
Level 1 — Object assessment. Estimation and
prediction of (ideally) single entity states on the
basis of inferences from observations. A frequent
example is the estimation of the kinematic
properties and the identification or classification of single objects of interest in a situation.
Level 2 — Situation assessment. Estimation and
prediction of entity states on the basis of inferred relations among entities. For a variety of
applications, this level strives to develop situation assessments based on the collective dynamics
among entities as well as the estimated relationships among them.
Level 3 — Impact assessment. Estimation and
prediction of effects on situations of planned or
estimated/predicted actions by participants (for
example, assessing susceptibilities and vulnerabilities to estimated/predicted threat actions
given one’s own planned actions). In military
applications, a major concern is threat estimation
that involves estimating most likely or most
damaging potential of an adversary.
Level 4 — Process refinement (an element of
resource management). Adaptive data acquisition and processing to support mission objectives. A frequent focus of level 4 design and
development is for a sensor management functionality, where observational strategies that
From Kessler et al. (1992).