Cybersecurity Education and Research
Computer Science Department
Dept. 3315, 1000 E. University ave
Laramie, WY 82071
Email: mike.borowczak@uwyo.edu
Michelangelo is a project still in the inception phase that is seeking to bring self-learning, genetic learning algorithms to the distributed systems space. These self-designing systems will seek to emulate previous behavior as discovered in SHARKS research before attempting to generalize to creating more advanced and unresearched swarm systems. This project also has the goal of making the design outputs of machine learning systems human-readable through applied Inductive Program Synthesis on a decentral-first DSL.
The Secure, Heterogeneous, Autonomous, and Rotational Knowledge for Swarms (SHARKS) protocol investigates distributed algorithms for swarm movement patterns. The drones in a given swarm have restrained compute resources and little memory, so developing efficient behaviors is necessary for the agents to work within these restrictions. In addition to positioning behaviors, this project investigates safety protocols to protect the swarm from adversarial swarms and environmental obstacles.
Over the summer of 2021, REU researchers Duke Tran and Bridget Humphreys worked alongside CEDAR researchers in order to add another dimension to the SHARKS algorithm. The researchers learned about what UAV swarms are and how they are secured while also developing a custom experimentation framework for the 3D SHARKS protocol in the Unity game engine. The third dimension of movement adds many complications to the algorithm such as finding the ideal distance between agents on the sphere or what dimension should the agents prioritize when moving? While these questions will need to be answered with future research, this work with the REU researchers laid the groundwork to begin answering those questions.
As more devices are added to peer-to-peer (P2P) networks in highly specialized, lightweight, and localized environments, the security and isolation of communication within a network remain a major design challenge. By examining several communication problems in specific domain applications, a secure communication protocol, secure communication for heterogeneous Internet of Things (IoT) with round-robin protection (SCHIRP), is used to explore and define potential minimalistic solutions.
Cybersecurity Education and Research
Computer Science Department
Dept. 3315, 1000 E. University ave
Laramie, WY 82071
Email: mike.borowczak@uwyo.edu