In the last decade, there has been a widespread deployment of autonomous vehicles for visual surveillance in civilian as well as military applications. In this talk, I will present a game-theoretic perspective to the problem of control and motion planning in a team of autonomous agents deployed in vision-based target tracking. In the first part of the talk, I will present a systematic development of a motion planner from locally optimal controllers for a single mobile agent that tries to track a mobile target in an environment containing obstacles. In the second part of the talk, I will present deployment, tracking and coordination strategies for multiple observers with guaranteed tracking performance by drawing a connection between multi-robot tracking and multi-robot coverage. The relevant tools are based on ideas from differential game theory and computational geometry. I will conclude my talk with some future directions and practical challenges in multi-robot target tracking and related problems.
Sourabh Bhattacharya received his MS in Electrical Engineering, MA in Applied Mathematics, and PhD in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign in 2010. After a postdoc at Coordinated Science Lab at UIUC, he joined Iowa State University as an Assistant Professor in the department of Mechanical Engineering. His research interests are in game theory and motion planning for multi-robot systems.