In many emerging applications, including autonomous driving, teleoperation, and rehabilitation robotics, a human and an automation are both responsible for controlling the system. In these scenarios, it is desirable that the system behaves close to what the human intends while providing high performance and safety. Shared autonomy is a general framework that deals with this problem by finding the right balance between applying the human's decisions and the automation's decisions. This talk will present some of our recent algorithmic frameworks on understanding human's unknown intent and preferences to control autonomous systems accordingly. We propose both offline and online controller synthesis and consider different forms of human feedback, including brain signals, intermittent force input, and selection. Finally, I will conclude this talk by discussing several ongoing and future research directions that we are pursuing to address the challenges of having competent, autonomous systems that interact with humans and empower them.

Mahsa Ghasemi is an Assistant Professor of Electrical and Computer Engineering at Purdue University. Her research focuses on theoretical and foundational advancements preparing autonomous systems to co-exist with humans in our complex world. Her contributions enable efficient and reliable integration of autonomy in various applications such as robotics, shared autonomy, and networked systems. Mahsa has been selected as a Rising Star in Electrical Engineering & Computer Sciences and a Rising Star in Machine Learning. She received her B.Sc. degree in Mechanical Engineering from Sharif University of Technology, and her M.S.E. and Ph.D. degrees in Mechanical Engineering and Electrical and Computer Engineering, respectively, from The University of Texas at Austin, where she was a recipient of the Cockrell School of Engineering Fellowship.

Host

Professor Sumeet Kumar Gupta, guptask@purdue.edu