Ravi Pandya

PhD Candidate

Carnegie Mellon University

I am a Ph.D. Candidate in the Robotics Institute at Carnegie Mellon University advised by Prof. Changliu Liu and Prof. Andrea Bajcsy. I am grateful to be funded by the NSF Graduate Research Fellowship. My PhD thesis is on using data-driven methods (e.g. reinforcement learning, trajectory forecasting, LLMs) to enable robots to safely interact with humans while accounting for the influence they have on peoples’ actions and intentions.

Prior to joining CMU, I worked as a Data Scientist at Ericsson where I used ML and multiagent reinforcement learning for optimizing radio networks. I also previously did research in human-robot interaction at UC Berkeley where I primarily worked with Prof. Anca Dragan, but also had the privilege of working with Prof. Ruzena Bajcsy.

📣 I will graduate in Summer 2025 and am actively seeking Research Scientist/Engineer positions in industry related to AI, ML, and robotics. Please reach out if you think I would be a good fit!

Interests

Machine Learning
Artificial Intelligence
Human-Robot Interaction
Optimal Control

Education

PhD in Robotics, 2020 - Present

Carnegie Mellon University
BS in Electrical Engineering and Computer Science, 2015-2019

UC Berkeley

Publications

A Unified Framework for Robots that Influence Humans over Long-Term Interaction

(in submission) arXiv preprint 2025.

We formulate a MOMDP (Mixed Observability Markov Decision Process) that accounts for the influence that robots have on humans in both short-term and long-term interactions. We show that existing approaches to influence are simplifications of our overall Unified framework that ignore the long-term influence on humans. Through simulations and user studies, we show that the solution to our unified framework results in influential long-term robot policies.

S. Sagheb, S. Parekh, Ravi Pandya, Y. Mun, K. Driggs-Campbell, A. Bajcsy, D.P. Losey

PDF Video

Robots that Learn to Safely Influence via Prediction-Informed Reach-Avoid Dynamic Games

International Conference on Robotics and Automation (ICRA) 2025.

We formulate a prediction-informed robust dynamic game to allow a robot to safely influence a human partner. We instantiate our method, called SLIDE (Safely Leveraging Influence in Dynamic Environments), in a simulated human-robot collaborative task. We find that SLIDE consistently enables the robot to leverage the influence it has on the human when it is safe to do so, ultimately allowing the robot to be less conservative while still ensuring a high safety rate.

Ravi Pandya, C. Liu, A. Bajcsy

PDF Website

Robust Safe Control with Multi-Modal Uncertainty

(in submission) arXiv preprint 2024.

We introduce a least-conservative robust safe controller for dynamical systems with additive and multiplicative multimodal uncertainty for energy-function-based safe control methods. We test our method on a simulated segway robot and find it is less conservative than existing unimodal robust control methods.

T. Wei, L. Ma, Ravi Pandya, C. Liu

PDF

Multimodal Safe Control for Human-Robot Interaction

American Control Conference (ACC) 2024.

We derive a least-conservative robust safe controller for dynamical systems with additive multimodal uncertainty (where additive refers to how the uncertainty enters into the dynamics with respect to the control input). We test our controller on a simulated human-robot system where the robot is uncertain of the human’s goal and find this approach to be safer than existing maximum-likelihood-based unimodal robust controllers.

Ravi Pandya, T. Wei, C. Liu

PDF

Multi-Agent Strategy Explanations for Human-Robot Collaboration

International Conference on Robotics and Automation (ICRA) 2024.

We introduce a novel method for generating explanations of collaborative strategies for humans and robots in tasks with multiple Nash equilibria. We generate a visual state-based explanation of what each agent should do in an upcoming collaboration. Ultimately, we find that our explanations help real participants better explore the full space of strategies and collaborate with autonomous partners more quickly.

Ravi Pandya*, M. Zhao*, C. Liu, R. Simmons, H. Admoni

PDF Video

Towards Proactive Safe Human-Robot Collaborations via Data-Efficient Conditional Behavior Prediction

International Conference on Robotics and Automation (ICRA) 2024.

We formulate a novel modification to typical human intention prediction via Bayesian inference that accounts for the influence that the robot will have on the person. Using this conditional behavior prediction model, the robot can proactively influence a human collaborator to choose efficient actions for the task. We find in a user study that participants tend to enjoy collaborating with this algorithm over baselines.

Ravi Pandya*, Z. Wang*, Y. Nakahira, C. Liu

PDF Video

Safe and Efficient Exploration of Human Models During Human-Robot Interaction

International Conference on Intelligent Robots and Systems (IROS) 2022.

We study the problem of adapting a robot’s dynamics model of a human collaborator online while staying safe; we test out controllers with different risk preferences and measure how they are affected by the presence of safe control. Ultimately, we find that a risk-seeking control can learn a good model, but necessitates activating the safety controller more than other methods.

Ravi Pandya, C. Liu

PDF Video

Nonverbal Robot Feedback for Human Teachers

(Oral, Acceptance 5.3%) Conference on Robot Learning (CoRL) 2019.

We study the problem of enabling a robot learner to give nonverbal feedback to a human teacher. We focus on using gaze as a predictor of the human teacher’s next action and find in simulation that this approach leads to faster and more accurate task learning. In both online and in-person user studies, we find that this nonverbal feedback also helps real human teachers get a better mental model of the robot learner and helps improve the robot’s learning performance.

Ravi Pandya*, S.H. Huang*, I. Huang*, A.D. Dragan

PDF Video

Human-AI Learning Performance in Multi-Armed Bandits

AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES) 2019.

We study how an AI agent can assist a human by suggesting options in a multi-armed bandit problem when both agents are learning the reward from arms from scratch. We find in a user study that people have two main modes of selecting arms that can be distibguished by the entropy of the arm frequencies over time, and that participants matched with an assistant with similar entropy profiles will be most helpful to them.

Ravi Pandya, S.H. Huang, D. Hadfield-Menell, A.D. Dragan

PDF DOI

Learning Image-Conditioned Dynamics Models for Under-actuated Legged Millirobots

(Best Paper Finalist) International Conference on Intelligent Robots and Systems (IROS) 2018.

We enable a small underactuated robot to learn how to walk on different terrains with a small amount of data collected in the real world by training a neural network dynamics model and running MPC over it to track trajectories. Importantly, the dynamics model takes in images of the environment to condition on, allowing the robot to learn different gaits for different terrains with just a single model.

A. Nagabandi, G. Yang, T. Asmar, Ravi Pandya, G. Kahn, S. Levine, R.S. Fearing

PDF Website DOI

Learning Human Ergonomic Preferences for Handovers

International Conference on Robotics and Automation (ICRA) 2018.

We focus on understanding how to best learn ergonomic preferences from a human in object handovers, since each person will have individual comfort preferences or constraints. We study an active learning approach to learning a human ergonomic cost function compared to passive and random baselines, and find that while active learning estimates the human’s cost function quickly, it incurs a higher ergonomic cost during learning.

A. Bestick, Ravi Pandya, R. Bajcsy, A.D. Dragan

PDF DOI

Ravi Pandya

PhD Candidate

Carnegie Mellon University

Interests

Education

News

Publications

A Unified Framework for Robots that Influence Humans over Long-Term Interaction

Robots that Learn to Safely Influence via Prediction-Informed Reach-Avoid Dynamic Games

Robust Safe Control with Multi-Modal Uncertainty

Multimodal Safe Control for Human-Robot Interaction

Multi-Agent Strategy Explanations for Human-Robot Collaboration

Towards Proactive Safe Human-Robot Collaborations via Data-Efficient Conditional Behavior Prediction

Safe and Efficient Exploration of Human Models During Human-Robot Interaction

Nonverbal Robot Feedback for Human Teachers

Human-AI Learning Performance in Multi-Armed Bandits

Learning Image-Conditioned Dynamics Models for Under-actuated Legged Millirobots

Learning Human Ergonomic Preferences for Handovers

Talks

Towards Influence-Aware Safe Human-Robot Interaction

Towards Influence-Aware Safe Human-Robot Interaction

Safely Influencing Humans in Human-Robot Interaction

Nonverbal Robot Feedback for Human Teachers

Contact