Multi-Domain Information Analytics (MDIA)

Abstract

This effort develops Artificial Intelligence/Machine Learning (AI/ML) approaches for providing Situational Awareness (SA) across echelons that are robust to compromised, corrupted, or limited data and networks in contested and unpredictable battlespace environments. These approaches will provide increased probability of discernment of true vs. false targets, and incorporate uncertainty-aware neuro-symbolic AI/ML to calibrate confidence in algorithm predictions. Research will incorporate multimodal analysis with multi-view scene understanding from heterogeneous sensor systems for context-aware inference, utilize transfer learning techniques to bridge domain gap between real and synthetic data for improved machine learning, and employ Size, Weight and Power-Time (SWaP-T) constrained processing at the edge on emerging low power secure compute architectures through neural network pruning and compression. Simulations of Command and Control (C2) strategies will incorporate the MDIA approaches.

Document Details

Document Type
Accomplishment
Publication Date
Oct 01, 2024
Source ID
81ae2a0398634fe257e26de049565ca8

Tags

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Computer Vision.
  • Enterprise Information Systems Architecture and Joint Command Capability Interoperability Support.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • AI & ML - Neural Networks
  • Fully Networked C3
  • Fully Networked C3 - Command and Control

Related Documents