Distributed Coordination Algorithms for Target Identification, Localization and Tracking with Mobile Visual Sensor Networks

Target tracking has been an active research area. It is now possible to use a large number of low-cost, autonomous mobile robots, as a single entity, for target identification, localization and tracking purposes. But, achieving optimal outcomes in a fully-distributed manner, without centralized control, is a difficult and still an open problem.

In this project, we propose to create fully-distributed algorithms that will guide a mobile visual sensor network to identify, localize and track targets in a given region. The locations of the targets are not known a priori. The mobile sensors scan a region to detect a target, once detected, some of the sensors get into a formation around the target for continuous monitoring, while the others search for more targets. To ensure scalable and robust operation, only local information is used. We will particularly focus on

The goal is to achieve optimal distribution of the sensors among the targets. We will work on theoretical development, mathematical analysis as well as empirical studies by implementing the algorithms on Monash University's Wireless Sensor and Robot Network Laboratory's experimental testbed.
Department of Electrical and Computer Systems Engineering, Monash University, Melbourne, Australia
Last modified: Sun Sep 14 18:20:46 EST 2014 by Ahmet Sekercioglu