Interactive Augmentation of Computer Generated Force Behavior Based on Cooperative and Reinforcement Learning. Phase 1.
Abstract
State-of-the-art Computer Generated Forces (CGFs) are predictable, non-adaptive, behaviorally distinguishable from manned simulators, and once "figured out," can be outsmarted by manned simulator crews. KATrix's NeurRule Technology integrates neural networks with rule-based systems, enabling SAF systems to continuously learn from manned simulators, human instructors, and its own mistakes. This Technology will make CGF behaviors more realistic, and will save time and money by permitting Battle Trainers to modify CGF behavior without the intervention of programmers. Phase I of this project has produced a design methodology for augmenting computer generated force behavior with the NeurRule Technology concepts of cooperative and reinforcement learning. The Phase I results indicate that (1) Intelligent CGFs can improve task performance through on-line learning, utilizing information from both supportive and adversarial sources, and (2) the NeurRulem Technology can coexist with existing SAF software such as modSAF and SAFDI. Phase II will extend these results by enabling SAF operators to interactively develop, through use of an appropriate graphical user interface, COF behaviors specific to dismounted-infantry operations in urban environments.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1995
- Accession Number
- ADA300346
Entities
People
- David A. Handelman
- Stephen H. Lane