Yunlong Song - 宋运龙

I am a Founding Research Scientist at Genesis AI.

I completed my PhD under the supervision of Prof. Davide Scaramuzza at the Robotics and Perception Group, part of the Department of Informatics and Institute of Neuroinformatics (University of Zurich & ETH Zurich). During my PhD, I was an associated researcher at the ETH AI Center and a visiting researcher at MIT. My work has involved close collaboration with leading researchers and institutions to tackle fundamental problems in AI and robotics. Highlights include developing a superhuman car racing agent with Sony AI; advancing autonomous drone racing in collaboration with Matthias Müller and Vladlen Koltun; building a differentiable simulation framework for quadruped locomotion with Prof. Sangbae Kim; exploring perceptive quadruped control with Prof. Fan Shi and Prof. Marco Hutter; and investigating fundamental reinforcement learning principles with Boris Belousov and Prof. Jan Peters.

Research Interests

My research focuses on developing next-generation robotic systems capable of learning, adapting, and understanding in complex real-world environments. I have explored various topics in robotics and machine learning, including optimal control, reinforcement learning, differentiable simulation, representation learning, and vision-based control. Beyond theoretical understanding, I have hands-on experience with a range of robotic platforms, including the Furuta Pendulum, Barrett WAM Arm, high-performance FPV drone, MIT Mini Cheetah, and robotic manipulators. My past research has primarily focused on developing robust skill-level policies for diverse robotic tasks. Currently, I am particularly interested in integrating large vision-language-action models with low-level robotic skills to enable autonomous systems to perform complex, real-world tasks with greater efficiency and adaptability. My current research interests include:

  • Large-scale Reinforcement Learning
  • Reinforcement Learning for Mobile/Whole-body Manipulation
  • Direct Policy Optimization for Pixel-to-Action Control
  • Post-training of Robotics Foundation Models
  • Multimodal Policy Learning

Selected Pulications

Some peronal favorites from my publication list:
  • Learning Vision-based Agile Flight via Differentiable Physics
    Y. Zhang*, Y. Hu*, Y. Song*, D. Zou and W. Lin
    Nature Machine Intelligence, 2025 PDF Video
  • Learning Quadruped Locomotion Using Differentiable Simulation
    Y. Song, S. Kim, and D. Scaramuzza
    CoRL (oral presentation), 2024 PDF Video
  • Reaching the Limit in Autonomous Racing: Optimal Control Versus Reinforcement Learning
    Y. Song, A. Romero, M. Müller, V. Koltun, and D. Scaramuzza
    Science Robotics (featured on the Cover), 2023 PDF Video
  • Policy Search for Model Predictive Control for Agile Drone Flight
    Y. Song and D. Scaramuzza
    IEEE T-RO, 2022 PDF Video
  • Flightmare: A Flexible Quadrotor Simulator
    Y. Song, S. Naji, E. Kaufmann, A. Loquercio, D. Scaramuzza
    CoRL, 2021 PDF Video
  • Learning High-level Policies for Model Predictive Control
    Y. Song and D. Scaramuzza
    IRSO, 2020 (my very first publication) PDF Video

Ph.D. Thesis