Kensuke Nakamura

ken-headshot.jpg

I am a Ph.D. student at Carnegie Mellon University’s Robotics Institute where I am advised by Prof. Andrea Bajcsy. My research leverages the synergy between optimal control and generative models to allow robots to safely operate in unstructured and uncertain environments. I develop theory and algorithms grounded in systems such as autonomous vehicles that use learned trajectory forecasters during planning or robotic manipulators that use world models to understand nuanced safety constraints. I was named a 2025 HRI Pioneer and I am fortunate to be supported by the NSF Graduate Research Fellowship.

I am currently an intern at NVIDIA with the Autonomous Systems and Physical AI Research (ASPIRE) Group led by Prof. Marco Pavone.

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

E-mail / Google Scholar / Github / Twitter

news

Sep 06, 2026 Happy to have been selected as one of eight Rising Stars at NERC 2026!
Sep 04, 2026 How Well Do Latent World Models Understand Partially Observable Safety Constraints? got accepted to CoRL 2026!
Jul 27, 2026 Gave talks at the Icon Lab and Hybrid Systems Lab at UC Berkeley, and the SISL lab and MSL lab at Stanford, where I went on a deep dive discussing “when” and “how” world models can assist in robot safety.
Jul 13, 2026 Happy to have co-organized the Rethinking What It Means to be Safe For Generalist Robot Workshop at RSS 2026 this year!
Feb 11, 2026 I gave a talk to the RobIn lab at UT Austin! I talked about fundamentals of safety-critical control (Hamilton-Jacobi Reachability and Control Barrier Functions) and two works that extended these concepts to latent safety filters.

selected publications

  1. L4DC
    nakamura_l4dc2026.png
    How to Train Your Latent Control Barrier Function: Smooth Safety Filtering Under Hard-to-Model Constraints
    Kensuke Nakamura, Arun L. Bishop, Steven Man, and 3 more authors
    In Learning for Dynamics and Control Conference, 2026
    arXiv
  2. CoRL
    multisafe_overview.png
    How Well Do Latent World Models Understand Partially Observable Safety Constraints?
    Matthew Kim, Kensuke Nakamura, and Andrea Bajcsy
    In Conference on Robot Learning, 2026
    arXiv
  3. ICRA
    anysafe_overview.png
    AnySafe: Adapting Latent Safety Filters at Runtime via Safety Constraint Parameterization in the Latent Space
    Sankalp Agrawal, Junwon Seo, Kensuke Nakamura, and 2 more authors
    In IEEE International Conference on Robotics and Automation (ICRA), 2026
    arXiv
  4. CoRL
    UNISafe_preview.gif
    Uncertainty-aware Latent Safety Filters for Avoiding Out-of-Distribution Failures
    Junwon Seo, Kensuke Nakamura, and Andrea Bajcsy
    In Conference on Robot Learning, 2025
    arXiv
  5. RSS
    frontfig_latentsafety3.png
    Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysis
    Kensuke Nakamura, Lasse Peters, and Andrea Bajcsy
    In Robotics: Science and Systems, 2025
    arXiv
  6. CoRL
    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
  7. CoRL
    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
  8. CDC
    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
  9. ICRA
    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