Online Modeling of Heterogeneous Autonomy
Abstract
Predicting an autonomous agent actions (human or articial) is an extremely important problem for autonomy research.Most existing methods for predicting an autonomous agent s action are online, i.e. training a model is done via batch based learning with the key assumption that the entire training data set is available prior to the learning task. Therefore existing methods are not suitable for emerging on line and mission critical applications with extensive human machine interaction and limited inter agent communication. In this research project we propose to address this gap by developing novel online learning techniques for predicting an intelligent agent s dynamic decisions.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 14, 2022
- Source ID
- FA95501910347
Entities
People
- Alfredo Garcia
Organizations
- Air Force Office of Scientific Research
- Texas Engineering Experiment Station
- United States Air Force