Kensuke Nakamura

prof_pic.jpg

I am a Ph.D. student at the Carnegie Mellon University Robotics Institute where I am advised by Prof. Andrea Bajcsy. My research focuses on safety for human-robot interaction. In particular, I’m interested maintaining rigorous guarantees drawn from control theory while reducing overconservativness by allowing robots to learn at runtime. I am fortunate to be supported by the NSF Graduate Research Fellowship.

Previously, I graduated from Princeton University where I was advised by Jaime Fernández Fisac and Naomi Ehrich Leonard. I’ve also had the pleasure of collaborating with Somil Bansal.

E-mail / Google Scholar / Github / Twitter

news

Sep 5, 2024 Our paper which derives a generalized regret metric to automatically extract system-level trajectory prediction failures was just accepted to the 2024 Conference on Robot Learning!
Jul 19, 2024 Deception Game won the Best Presentation Award at the RSS ‘24 Safe Autonomy workshop!
Apr 4, 2024 I was awarded the NSF GRFP!
Mar 7, 2024 I submitted my first paper as PhD student at CMU! We derived a new regret metric that automatically identifies prediction errors that induced system-level failures during deployment. Finetuning with only this high-regret data leads to large system-level performance improvements despite only using a small fraction of the total deployment data.
Aug 30, 2023 Our paper on adaptive safety guarantees that are aware of the robot’s learning dynamics has been accepted to the 2023 Conference on Robot Learning!

selected publications

2024

  1. qualitative-regret.png
    Not All Errors Are Made Equal: A Regret Metric for Detecting System-level Trajectory Prediction Failures
    Kensuke Nakamura, Ran Tian, and Andrea Bajcsy
    In 8th Annual Conference on Robot Learning, 2024
    arXiv

2023

  1. framework_v2.png
    Deception Game: Closing the Safety-Learning Loop in Interactive Robot Autonomy
    Haimin Hu, Zixu Zhang, Kensuke Nakamura, and 2 more authors
    In 7th Annual Conference on Robot Learning, 2023
    arXiv
  2. emergent.png
    Emergent Coordination through Game-Induced Nonlinear Opinion Dynamics
    Haimin Hu, Kensuke Nakamura, Kai-Chieh Hsu, and 2 more authors
    In 2023 62nd IEEE Conference on Decision and Control (CDC), 2023
    arXiv
  3. confidence.gif
    Online Update of Safety Assurances Using Confidence-Based Predictions
    Kensuke Nakamura, and Somil Bansal
    In 2023 International Conference on Robotics and Automation (ICRA), 2023
    arXiv