美诱直播

ABSTRACT

TRACER works alongside the human planner to assist with the dynamic targeting prosecution challenge: severely compressed decision cycles where the human?planner must identify available all-domain assets with the right capabilities and?perform cost/benefit assessment in time to achieve a desired effect.

TRACER provides a set of optimized solutions at machine speed, each with a?primary objective and set of weighted constraints. It minimizes the rippling effect?of re-tasking while maximizing change tolerance in the operational environment.

?

WHY?

TRACER maximizes human planning expertise, while?minimizing operator burden, by doing the “heavy lifting”?of data analysis, which allows human planners to focus?on decisions computers cannot and should not make.

?

WHAT?

By building on prior planning systems, TRACER enables?all-domain dynamic target (DT) prosecution that scales to?hundreds of DTs during a 24-hour planning cycle. It places?the human planner in the loop, allowing them to tailor?the automation level and visualize option generation?rationale, thus promoting system trust.

?

HOW?

The TRACER system is a research prototype designed to demonstrate the?feasibility, advantages, and limitations of a DT decision support tool, with?a desired level of maturity at the end of Phase 2. It is composed of several?microservices, which builds on prior work.

The Multi-Objective Service allows the human planner to weigh their?preferences over a set of planning objectives, feeding into the Option?Generation Service that generates operationally relevant Effects/Integrated?Tasking Order (E/I-TO) change options for prosecuting multiple DTs during?plan execution. The Resource-Task Pairing Service uses an auction-based?allocation engine to assign all-domain resources to these options while?observing dependencies and minimizing ripple-effect changes. The Cost/Benefit Analysis Service provides visualizations to explain option generation?and trade-offs across a set of factors.

?