Title: Proactivity and Personalization in Longitudinal Robotic Assistance

Date: Wednesday, June 18th, 2025

Time: 1:30pm - 3:30pm ET

Location: Coda C1115 Druid Hills, or  Zoom Link

Maithili Patel

Robotics Ph.D. Student

School of Interactive Computing

Georgia Institute of Technology

Committee:

• Dr. Sonia Chernova (advisor) – School of Interactive Computing, Georgia Institute of Technology
• Dr. Matthew Gombolay – School of Interactive Computing, Georgia Institute of Technology
• Dr. Mark Riedl – School of Interactive Computing, Georgia Institute of Technology
• Dr. Roberto Martín-Martín  – School of Computer Science, University of Texas, Austin
• Dr. Jacob Andreas – Electrical Engineering and Computer Science, Massachusetts Institute of Technology

Abstract:

Robots today are evolving towards versatile agents that can assist humans in daily lives over long periods of time. For robots to be effective general-purpose assistants, they must seamlessly integrate into users’ daily lives, and proactively attend to user needs over extended interactions. To function as truly proactive assistants, robotic systems must have the following characteristics: adaptation to the dynamic nature of human environments and variability in tasks that humans perform, anticipation of user behavior to provide timely assistance without needing to be asked, and personalization of the mode of assistance and manner of task execution to align with the user’s unique preferences.

In this thesis, I contribute computational methods that enable robotic systems to develop context about the users’ routines and preferences, and leverage such context to enable proactive behaviors over longitudinal interactions. My work utilizes unobtrusive observations and limited feedback obtained over time to develop an understanding of the users’ routine lives, ensuring minimum disturbance to the user.