Reliable Interactive Autonomy Lab

We are the Reliable Interactive Autonomy (RIA) Lab in the Luddy School of Informatics, Computing, and Engineering at Indiana University Bloomington.

We build robots and AI agents that learn from the people they work with, know what they don’t know, and collaborate reliably over long horizons. Our work sits at the intersection of interactive machine learning, uncertainty quantification, and human-robot interaction.

Three questions drive us:

• How should a robot ask? Human time and attention are scarce. We design algorithms that decide when to query a person, what to ask, and when to just act, so that a robot learning on the job never overburdens its partner.
• How should a robot know when it is wrong? We give learned policies statistically rigorous notions of uncertainty, so a robot can recognize the moment it has left familiar territory and hand control back before it fails.
• How should a robot and a person align? Collaboration is bidirectional. We study how robots can infer human strategies and preferences, and how they can explain their own behavior so people can predict and trust them.

Our long-term goal is lifelong, collaborative intelligence: robotic partners that keep learning through strategic interaction with the people around them, across tasks and across years.

We are recruiting!

Our first PhD students join in Fall 2026, and we are looking for more PhD students, Master’s students, and undergraduates who are excited about robotics, interactive machine learning, and human-AI interaction. See how to join us for details.