Tianyu Qiu

Ph.D. Student @ UT Austin

CLeAR Lab

The University of Texas at Austin

Biography

Tianyu Qiu is currently a Ph.D. student in Aerospace Engineering in the University of Texas at Austin. He is also a research assitant in the Control and Learning for Autonomous Robotics (CLeAR Lab), advised by Prof. David Fridovich-Keil. Prior to UT Austin, Tianyu received his M.Eng. and B.Eng. degree from Shanghai Jiao Tong University in 2023 and 2020, respectively. Tianyu’s research interests include reinforcement learning, game theory, social navigation and SLAM.

Interests

Reinforcement Learning
Game Theory
Social Navigation
SLAM

Education

Ph.D. in Aerospace Engineering, 2028
The University of Texas at Austin, Austin, TX, USA
M.Eng. in Electronic Information, 2023
Shanghai Jiao Tong University, Shanghai, China
B.Eng. in Electrical & Computer Engineering, 2020
Shanghai Jiao Tong University, Shanghai, China
High School Diploma, 2016
Shanghai High School, Shanghai, China

News

[2024-10] Our paper "Inferring Occluded Agent Behavior in Dynamic Games from Noise Corrupted Observations" has been accpeted to Robotics & Automation Letters!

[2024-07] We submitted paper "Inferring Occluded Agent Behavior in Dynamic Games from Noise Corrupted Observations" to RA-L.

[2024-02] We submitted paper "Dense Dynamics-Aware Reward Synthesis: Integrating Prior Experience with Demonstrations" to ICML 2024.

[2023-03] I'm glad to announce that I've graduated with Master of Engineering degree in Electronic Information from Shanghai Jiao Tong University! Thank you all for the help in my seven years' journey!

[2023-02] I'm glad to annouce that I'm joining University of Texas at Austin as an Aerospace Engineering Ph.D. student in 2023 Fall. See you in Austin, TX!

Projects

Oct 22, 2024

Inferring Occluded Agent Behavior in Dynamic Games from Noise Corrupted Observations

Accepted by RA-L

Feb 1, 2024

Dense Dynamics-Aware Reward Synthesis: Integrating Prior Experience with Demonstrations

Submitted to ICML 24'

Mar 15, 2023

Pedestrian Trajectory Prediction and Mobile Robot Navigation Based on Inverse Dynamic Games

Master Thesis

Dec 25, 2019

Healthcare Telepresence Robot for the Elderly

Major Capstone Design

Featured Publications

Tianyu Qiu, David Fridovich-Keil

July, 2024 Submitted to RA-L

Inferring Occluded Agent Behavior in Dynamic Games with Noise-Corrupted Observations

This paper presents an occlusion-aware game-theoretic inference method to estimate the locations of potentially occluded agents, and simultaneously infer the intentions of both visible and occluded agents, which best accounts for the observations of visible agents. Additionally, we propose a receding horizon planning strategy based on an occlusion-aware contingency game designed to navigate in scenarios with potentially occluded agents.

Cevahir Koprulu, Po-han Li, Tianyu Qiu, Ruihan Zhao, David Fridovich-Keil, Sandeep Chinchali, Ufuk Topcu, Tyler Westenbroek

February, 2024 Submitted to ICML2024

Dense Dynamics-Aware Reward Synthesis: Integrating Prior Experience with Demonstrations

This work introduces a systematic reward-shaping framework which distills the information contained in 1) a task-agnostic prior data set and 2) a small number of task-specific expert demonstrations, and then uses these priors to synthesize dense dynamics-aware rewards for the given task. This supervision substantially accelerates learning in our experiments, and we provide analysis demonstrating how the approach can effectively guide online learning agents to faraway goals.