Improvement of Multi-Target Tracking in a Multi-Agent Architecture with Multi-Sensor Data Fusion

This Article presents a Surveillance Multi-Agent System (S-MAS) architecture which focuses on the fusion of data from multi sensors for enhanced automotive safety andtraffic efficiency. In S-MAS tools will be introduced as autonomous agents for implementing a multi-sensor data fusionat architectural level: surveillance–sensor agents, a fusion agent, interface agents, record agents, planning agents, etc. They differ in their ability to carry out a specific surveillancetask. A surveillance–sensor agent controls and manages individual sensors. In this work we focus on the fusion agent,addressing specific problems of on-line sensor alignment, registration, bias removal and data fusion. We show how the inclusion of this fusion agent guarantees that objects of interest are successfully tracked across the whole area.

Keywords:

Multi Agent Systems , Multi-Sensor Multi-Target Tracking , Random Sensor Data Fusion