PDE-based Analysis of Cyber-Attacks in Vehicle Swarms
Meysam Ghanavati, Animesh Chakravarthy, Prathyush P. Menon · 2018
This paper considers scenarios wherein a hacker hacks into a subset of UAVs in a swarm and turns them into vehicles with malicious intent. The objective of the malicious vehicles is to prevent the swarm from performing its intended mission, and thereby compromise the effectiveness of the overall swarm. The two species of UAVs in the swarm - malicious UAVs and normal UAVS, as well as their associated interaction effects, are modeled using Partial Differential Equations (PDEs). The malicious UAVs choose a reference density profile of the normal UAVs which, when enforced, would reduce the effectiveness of the swarm. The malicious UAVs then perform a sequence of velocity changes with the objective of changing the density profile of the normal UAVs so that it tracks the reference density profile. Analytical expressions governing the velocity inputs of the malicious vehicles with which they can generate such density changes in the normal UAVs, are demonstrated. Simulations are presented to demonstrate the working of the attack model.