*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: A Multi-layer Swarm Control Model for Information Propagation and Multi-tasking
Committee:
Dr. Fumin Zhang, ECE, Advisor , Advisor
Dr. Yorai Wardi, ECE
Dr. Samuel Coogan, ECE
Dr. Magnus Egerstedt, ECE
Dr. Molei Tao, Math
Abstract:
While individuals in natural swarms are collectively performing complex tasks such as foraging or synchronization, critical information such as predator warnings propagate across the swarm almost instantly and presumably without explicit communication between the individuals. In this Dissertation, we propose a novel multi-layer control model composed of an interplay of decentralized algorithms for perception and swarming. Through this model, we demonstrate implicit information propagation and multi-tasking in swarms using only local interactions and without explicit communication or prescribed formations. Additionally, we show how this bio-inspired model enabled us to design a multi-agent distributed control law to solve the source seeking and level curve tracking problems without relying on the explicit communication of the field measurements nor the estimation of the field gradient. The main difficulty this model is set to solve is how to detect critical information in motion behavior and then respond in a way that allows the information to propagate while at the same time performing a collective task. We overcome this difficulty by using PCA learning algorithms in the perception layer to capture the variations in the spatial distribution of the surrounding agents. Additionally, we design a control law that autonomously balances between responding to the local changes and synchronizing with the other individuals. This results in an agile behavior where the motion is stable enough to achieve a certain task but at the same time flexible enough to allow important information to propagate. We validate the efficiency of the proposed algorithms via various simulation and experimental results using the indoor flying Georgia Tech Miniature Blimps and the Georgia Tech Robotarium mobile robots. The proposed model has the potential to be used to design various swarm algorithms - especially those incorporating individual differences between agents - such as designing tactics for a swarm of drones to avoid or chase a malicious agent.
