Socially-Aware Autonomy: Game-Theoretic Planning and Control For Multi-Agent Interactions


To transform our lives, robots need to interact with humans and other robots in complex shared environments. For example, autonomous cars need to interact with pedestrians, human-driven cars, and other autonomous cars, while autonomous delivery drones need to interact with drones and humans. A core difficulty in building such interactive robots is accounting for the dependence of all agents' decisions upon one another. In such interactions, robots must be cognizant of their influence on other agents and society in general. This talk presents an overview of how we can leverage game-theoretic planning and control to account for agents’ interactions and develop socially-aware autonomy. I will discuss the opportunities and challenges associated with game-theoretic planning across various scales of interactions. I will start from motion planning and then consider higher-level robot decision making in interactive settings. Finally, I will focus on the societal implications of robotic interactions.


Negar Mehr is an assistant professor at the UIUC Aerospace Engineering department. Previously, she was a postdoctoral scholar at Stanford Aeronautics and Astronautics department. She received her PhD in the Mechanical Engineering department at UC Berkeley in 2019 and received her B.Sc. in Mechanical Engineering from Sharif University of Technology, Tehran, Iran, in 2013. Her research interests lie at the intersection of control theory, robotics, and game theory.Specifically, she is interested in developing control algorithms that enable robots to safely and intelligently interact with each other and humans. Negar was the recipient of the IEEE ITSS best PhD dissertation award in 2020. She won the best student paper award at the International Conference on Intelligent Transportation Systems, 2016, and was recognized as a rising star in EECS, Aeronautics & Astronautics, and Civil and Environmental Engineering.