Interpretable Multimodal Sensor Fusion

Abstract

Multimodal sensor fusion is concerned about integrating information from multiple sources which observe different physics laws and have different uncertainty models. The PI’s research group is investigating deep learning based dynamic multimodal sensor fusion. When using machine learning algorithms for sensor fusion, it is usually not transparent to the users how the algorithms derive information and make decisions. For mission critical applications, such as automatic target recognition and tracking (ATRT), interpretability and transparency of the sensor fusion system are important to support missions and deploy artificial intelligence models. We propose to investigate how to create transparency to the fusion process and provide insights into the effectiveness of different sensing modalities with respect to the decision-making process.

Document Details

Document Type
DoD Grant Award
Publication Date
Mar 07, 2023
Source ID
FA95502110224

Entities

People

  • Jia Li

Organizations

  • Air Force Office of Scientific Research
  • Oakland University
  • United States Air Force

Tags

Fields of Study

  • Computer science

Readers

  • Geospatial Intelligence and Artificial Intelligence Analytics
  • Neural Network Machine Learning.
  • Sensor Fusion and Tracking Systems.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy