Multi-Agent System for Tracking the Intent of Surface Contacts in Ports and Waterways

摘要

Maritime security is especially critical for countries like Singapore, an island nation situated on the world's busiest shipping routes, whose economic prosperity is highly dependent on international trade from her busy port, petrochemical complexes and other high value units located along her coastline. This thesis borrows the ideas and techniques suggested for identifying air threats in the Air Defense Laboratory (ADL) and employ them to identify asymmetric maritime threats in port and waterways. Each surface track is monitored by a compound multi-agent system that comprise of the several intent models, each containing a nested multi-agent system. The attributes that define intent models of friendly, neutral, unknown and potentially hostile surface contacts are obtained from movement and communication protocols defined by the Vessel Traffic Information System (VTIS), maritime navigation rules and cues for surface warfare threat assessment. The underlying cognitive mechanism of the models is conceptual blending. The study includes a simulation of a mock VTS for the port of Singapore and surrounding waterways to test the ability of the models to compress data and information regarding multiple simulated surface contacts into integration networks and then determine the surface contacts' intent through the expansion of the integration networks.