MnRI Seminar: Yasin Yazıcıoğlu
June 12, 2020 — 2:00PM CST — Zoom Meeting Link
"Distributed Learning for Optimal Planning in Persistent Cooperative Missions"
Teams of autonomous systems have proven to have great potential for serving as robust and efficient solutions in various applications such as environmental and infrastructure monitoring, search and rescue, precision agriculture, manufacturing, and logistics. Realizing this potential mainly hinges on achieving a proper coordination (an optimal joint plan) among the team, which usually leads to intractable large-scale combinatorial optimization problems. This intractability necessitates the design of scalable distributed control, learning, and optimization algorithms that can work under the partial information available to the agents and achieve performance guarantees (e.g., convergence to optimal plans).
In this talk, I will present a game-theoretic approach to distributed learning of optimal plans in long-horizon (persistent) missions involving cooperative tasks with time windows. Motivated by applications such as industrial/warehouse automation, precision agriculture, and environmental monitoring, we consider a setting where each task has a value and can be completed if sufficiently many robots attend to the task at the specified location within the specified time window. Tasks keep arriving periodically over episodes and the robots try to maximize the total value of completed tasks by planning their own trajectories in the upcoming episode based on their past observations. I will show how such a distributed planning problem can be solved with performance guarantees by decomposing the team (global) objective function into properly designed individual (local) utility functions and using game theoretic learning algorithms. Finally, I will conclude the talk with some ongoing work and future directions including the application of these methods to heterogeneous systems (e.g., human-robot teams) and systems under complex spatio-temporal constraints.
About Yasin Yazıcıoğlu
Yasin Yazıcıoğlu is a research assistant professor in the Department of Electrical and Computer Engineering and affiliated with the Minnesota Robotics Institute at the University of Minnesota. Prior to joining the University of Minnesota, he was a postdoctoral research associate in the Laboratory for Information and Decision Systems (LIDS) at MIT from 2014-2017.
He received the Ph.D. degree in Electrical and Computer Engineering from the Georgia Institute of Technology in 2014, and the B.S. and M.S. degrees in Mechatronics Engineering from Sabancı University, Turkey, in 2007 and 2009 respectively. His research is primarily focused on distributed control, optimization, and learning with applications to cyber-physical-societal systems, networks, and robotics.
