From Single-Robot Learning and Planning to Multi-Robot Decision Making

Date: 
Tuesday, March 13, 2018 - 11:00
Location: 
Blakeslee Room
Presenter: 
Pooyan Fazli
Abstract: 
Robots are gradually moving from factories and labs to streets, homes, offices, and healthcare facilities. Future robots will need to operate in dynamic, uncertain, adversarial, and multi-goal environments. In many real-work applications, even when a single robot can achieve a given task, the possibility of employing a team of robots can improve the overall performance and robustness of the system. Moreover, as robots are increasingly becoming part of the everyday lives of humans, it is essential for them to be aware of the surrounding humans while performing their tasks in the environment. In this talk, I will first introduce a probabilistic local planner for the safe navigation of an autonomous robot in dynamic and unknown environments and discuss an imitation learning-based approach to enable the robot to navigate while complying with social norms and ensuring human safety. Next, I will present a general framework that allows remote and heterogeneous robots to share plans and instructions on the tasks assigned to them. Lastly, I will focus on the multi-robot coverage problem and discuss a distributed learning algorithm to patrol and monitor adversarial environments. I will conclude my discussion with an overview of my ongoing and future research.
Bio: 

Pooyan Fazli is an assistant professor and the founding director of the People and Robots Laboratory (PeRL) in the Electrical Engineering and Computer Science Department at Cleveland State University (CSU). He received his Ph.D. in computer science from the University of British Columbia. Prior to joining CSU, he was a postdoctoral fellow in the CORAL Research Group at Carnegie Mellon University and in the Laboratory for Computational Intelligence at the University of British Columbia. His research focuses on artificial intelligence, autonomous robots, multi-robot systems, human-robot teams, robot learning, and robot vision and perception.