Maritime smuggling detection and mitigation using risk-aware hybrid robotic sensor networks

Abstract

With the rise of more resourceful unmanned aerial vehicles (UAVs), their inclusion into robotic sensor networks (RSNs) is inevitable. The highly mobile nature of UAVs allows greater monitoring capabilities, making them most suitable for RSNs. Compared to traditional nodes in RSNs, UAVs suffer even more from communication disruptions and energy depletion, must often rapidly determine actions for themselves, and consequently require more autonomy. Prior work has been done in wireless sensor network (WSN)/aerial sensor network (ASN) coordination in a few applications such as protecting critical infrastructure, restoring communication between nodes, and healing networks, while other work has been accomplished on using the UAV network for augmenting the monitoring capabilities of WSNs. We introduce a novel methodology to integrate UAVs into RSNs for monitoring purposes by formulating the problem in the context of a risk management framework (RMF). This methodology allows a more precise risk feature classification and a more efficient task allocation for the ground network by utilizing the monitoring capabilities of the UAVs to informatively warn the RSN of any incoming events. We also present a fictitious but credible maritime smuggling scenario near the Port of Barcelona based on expert knowledge, and apply the methodology to detect and mitigate maritime smuggling. The network's behaviour is traced throughout the scenario and is repeated with civilian ships to assure that they are not flagged as smugglers. The applied methodology results in a successful classification and mitigation of the smuggling activity.