Kazuki Mizuta

Kazuki is a graduate student in the Aeronautics & Astronautics Department. His research revolves around the intersection of machine learning and control theory, with the ultimate goal of realizing autonomous mobile robots in uncertain and dynamic environments. His research interests are diverse and include:

  • Deep Generative Models for Autonomous Robots: Incorporating deep generative models into the autonomous robots to improve robot’s perception and decision-making abilities in complex real-world settings.
  • Safety-Critical Systems in Uncertain Environments: Combining control theory and machine learning to robustly guarantee the safety of control systems in uncertain and dynamic scenarios in real-time.
  • Cooperative Task Execution by Multiple Robots: Coordinating multiple robots to accomplish complex objectives efficiently and adjust their roles dynamically to adapt to changing conditions.

Kazuki is partially supported by the Nakajima Foundation.


Publications

  1. CoBL-Diffusion: Diffusion-Based Conditional Robot Planning in Dynamic Environments Using Control Barrier and Lyapunov Functions
    Mizuta, K., and Leung, K.
    In IEEE/RSJ Int. Conf. on Intelligent Robots & Systems, (preprint)
    Arxiv Bib
  2. CoBL-Diffusion: Diffusion-Based Conditional Robot Planning in Dynamic Environments Using Control Barrier and Lyapunov Functions
    Mizuta, K., and Leung, K.
    In Proc. IEEE Conf. on Robotics and Automation: Long-term Human Motion Prediction Workshop, 2024
    Bib

Related news