Target Re-Association for Autonomous Agents (TRAAA)
Abstract
Teams of autonomous agents able to understand (perceive) a physical environment and share suitableinformation towards a global common objective (e.g., detecting mission-relevant objects, entities,and actors in the environment, infer their functions, activities and intentions, and predict events) areof paramount importance for surveillance purposes within the Naval Domain.Achieving a common goal requires a team performing and sharing suitable representations both forperception and planning. This is often not the case of existing solutions, where the two tasks aregenerally incompatible, so joint architectures and algorithms for learning of unified representationsare needed for a better integration of perception and planning.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 11, 2020
- Source ID
- N629092012075
Entities
People
- Huanyu Cheng
Organizations
- Office of Naval Research
- United States Navy
- University of Udine