TY - GEN
T1 - Defensive Strategies for Autonomous Decentralized Lightweight Swarms
AU - Wolf, Shaya
AU - Borowczak, Mike
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11/9
Y1 - 2020/11/9
N2 - The SHARKS protocol enables agents to circle a two-dimensional target or ensphere a three-dimensional target through two simple distributed algorithms that ensure that the drones are equally dispersed from each other and maintain a specified distance from a target. An attack vector was identified in previous work that corrals agents and inhibits the swarms ability to create a perimeter. This research focuses on overcoming this attack without adding extensive overhead to the drones. Three basic techniques are proposed where the legitimate agents eject from their ideal circle around the target. By breaking out of this formation and quickly returning, the agents are able to bypass the adversarial nodes. The results were measured using a security metric based on the total disruption of the adversaries. These experiments showed that the total disruption across the swarm could be less than five percent using ejection strategies that set the ejection distance or the likelihood of ejecting based on the distance between the agents.
AB - The SHARKS protocol enables agents to circle a two-dimensional target or ensphere a three-dimensional target through two simple distributed algorithms that ensure that the drones are equally dispersed from each other and maintain a specified distance from a target. An attack vector was identified in previous work that corrals agents and inhibits the swarms ability to create a perimeter. This research focuses on overcoming this attack without adding extensive overhead to the drones. Three basic techniques are proposed where the legitimate agents eject from their ideal circle around the target. By breaking out of this formation and quickly returning, the agents are able to bypass the adversarial nodes. The results were measured using a security metric based on the total disruption of the adversaries. These experiments showed that the total disruption across the swarm could be less than five percent using ejection strategies that set the ejection distance or the likelihood of ejecting based on the distance between the agents.
KW - Autonomous Swarms
KW - Consumer Electronics
KW - Decentralized Control
KW - UAVs
UR - http://www.scopus.com/inward/record.url?scp=85101938116&partnerID=8YFLogxK
U2 - 10.1109/ICCE-Berlin50680.2020.9352161
DO - 10.1109/ICCE-Berlin50680.2020.9352161
M3 - Conference contribution
AN - SCOPUS:85101938116
T3 - IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin
BT - 2020 IEEE 10th International Conference on Consumer Electronics, ICCE-Berlin 2020
PB - IEEE Computer Society
T2 - 10th IEEE International Conference on Consumer Electronics, ICCE-Berlin 2020
Y2 - 9 November 2020 through 11 November 2020
ER -