Enhancing Simulation-based Training Adversary Tactics via Evolution (ESTATE)
Abstract
To extend modern simulation-based training environments to incorporate tailored simulation-based training experiences to facilitate accelerated skills development and sustainment, we are developing a system for Enhancing Simulation-based Training Adversary Tactics via Evolution (ESTATE). The system consists of an off-line adaptation engine to extract a model of the trainee based on past performance and generate tailored challenge problems for on-line training. Off-line adaptation is performed using evolutionary algorithms (EAs) to search through the space of challenges to exploit fundamental weaknesses in trainee strategy and tactics. The full-scope prototype ESTATE system is targeting simulation-based training systems within the Deployable Virtual Training Environment (DVTE) to support the squad-level training of U.S. Marines. This report documents the Option 1 period of the effort.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 15, 2010
- Accession Number
- ADA534778
Entities
People
- Bjorn Gunnarsson
- Brad R Rosenberg
- James Niehaus
- Jordan Pollack
- Max Metzger
- Scott N. Reilly