Maritime Domain Awareness via Agent Learning and Collaboration
Abstract
Maritime security is vital to US security. Enhanced Maritime Domain Awareness (MDA) of potential threats in this dynamic environment can be achieved, yet requires integrated analysis from numerous sources in real time. We will present a learning agent technology that integrates structured and unstructured data and discovers behavior patterns from varied sources such as Automatic Information Systems (AIS), Coast Guard, and police contextual information including: maritime commercial activities, weather, terrain, environmental conditions, maritime incidents, casualties, and military exercises. These discovered patterns can help correlate warnings and reduce false alarms in support of maritime security. We will show our test results from the Trident Warrior (TW08) exercise. We will also discuss the agent learning applied to system self-awareness, where we consider that the cognitive interface between decision makers and a complex system may be expressed in a range of terms or "features," i.e. specific vocabulary to describe a System of Systems (SoS) or so-called Lexical Link Analysis (LLA). MDA is an extremely varied and dynamic SoS, requiring constant collaboration and decision making. We will discuss prototypes of agent learning and collaboration, LLA, and visualization that provide real-time "views" of SoS to support large-scale decision making for MDA technology acquisition, irregular warfare at sea, and intelligence collection with analysis automation.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 24, 2010
- Accession Number
- ADA525250
Entities
People
- Charles Zhou
- Douglas J. MacKinnon
- Shelly P. Gallup
- Ying Zhao
Organizations
- Naval Postgraduate School