Abstract

The focus of this thesis is directed towards developing distributed coordination protocols for a group of agents (mobile sensors) deployed to cover an area of interest. It is assumed that sensors are subject to an alert message at any point in time, which is issued in an emergency situation. Such an emergency event is formulated as an abrupt change in the coverage priority of specific regions in the field. In the normal situation, a protocol is used to move the mobile sensors in the plane in such a way that the overall sensing coverage is increased. Then, as soon as an alert message is issued, sensors that receive the information communicate with their neighbors to inform them of the message. An appropriate number of sensors are subsequently tasked to further improve the coverage of the specified area by adjusting their positions iteratively to increase the coverage of the alert area. Two types of algorithms are developed, where the first one is mainly focused on the alert area coverage, and the second one aims to also cover the rest of the field as much as possible. The algorithms are Voronoi-based, and are guaranteed to increase the desired sensing coverage at each iteration. Some examples and comparative results are provided to demonstrate the effectiveness of the proposed algorithms.