Tools for Real-Time Anticipation of Enemy Actions in Tactical Ground Operations
Abstract
DARPA has recently undertaken a research project titled Real-time Adversarial Intelligence and Decision-making (RAID), which provides in-execution predictive analysis of probable enemy actions. A particular focus of the program is tactical urban operations against irregular combatants -- an especially challenging and operationally relevant domain. The RAID program leverages novel approximate game-theoretic and deception-sensitive algorithms to provide real-time enemy estimates to a tactical commander. In doing so, the RAID program is addressing two critical technical challenges: (a) adversarial reasoning: the ability to continuously identify and update predictions of likely enemy actions; (b) deception reasoning: the ability to continuously detect likely deceptions in the available battlefield information. Realistic experimentation and evaluation is driving the development process using human-in-the-loop, wargames to compare humans and the RAID system. This paper provides a discussion of the techniques and technologies chosen to perform the adversarial and deception reasoning. It also provides details about the experiments and experimentation environment that are used to demonstrate and prove the research goals.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2005
- Accession Number
- ADA460912
Entities
People
- Alexander Kott
- Michael Ownby
Organizations
- Defense Advanced Research Projects Agency