Mobile-Agent-Based Autonomous Data Fusion for Distributed Sensors

Impact Technologies, LLC in collaboration with Wright State University and Northrop Grumman, proposes to develop and demonstrate a mobile-agent-based decentralized data fusion system that fuses the data from geographically-dispersed heterogeneous sensors into an integrated ISR information system. Unlike conventional centralized methods, in which each sensor sends local information to a processing center for integration, in the proposed approach, data will first be processed on the sensor nodes, then fused locally within a group of unmanned vehicles using lightweight mobile-agents before the information is further fused at system level. Instead of transmitting large amounts of data over the network, the mobile-agent with partially integrated results migrates from one node to another performing data fusion at each node visited. This results in a reliable power-efficient data fusion process that improves decision accuracy in a progressive fashion without requiring a high bandwidth. The system will be built on a hierarchical open system architecture using standard protocols while the software implementation and integration will follow a well-established system engineering approach. At the conclusion of Phase I, a design concept will be developed and feasibility will be demonstrated through software simulation using typical littoral ISR scenarios.