Scott Smolka 'A Bird's-Eye View of Flocking'

Dates: 
Friday, April 2, 2021 - 4:00pm to 5:00pm
Location: 
Zoom - contact events@cs.stonybrook.edu for Zoom info.
Event Description: 

Abstract:
In this talk, I will discuss recent advances in computational approaches
to flocking behavior in multi-agent systems.  The talk will be in three
parts.  In the first part, I will present Declarative Flocking (DF), our
approach to the flocking problem based on optimal control, Model Predictive
Control in particular.  DF allows us to capture various aspects of flocking
behavior (basic flocking, target seeking, predator avoidance, etc.) simply
as terms of a cost function.  In the second part, I will present our work
on Neural Flocking (NF) which demonstrates how a symmetric and fully
distributed flocking controller can be synthesized using Deep Learning
from a centralized flocking controller.  Finally, I will discuss how we
extend Declarative Flocking to capture high-speed, flock-wide turning
maneuvers, a phenomenon seen in starling flocks.

Time permitting, I will also give an overview of work we are doing in
assuring the safety and reliability of networked microgrids, a rapidly
emerging trend in power engineering.

Bio:
Scott A. Smolka
SUNY Distinguished Professor
Department of Computer Science
Stony Brook University

Computed Event Type: 
Mis