Portrait of Qianzhong Chen
Robotics · Embodied AI · Stanford

Qianzhong Chen 陈钱中

PhD Student in Aeronautics & Astronautics at Stanford University

I am a PhD student advised by Mac Schwager. My goal is to build general-purpose robots that can perform complex manipulation tasks in homes and factories.

My research spans vision-language-action models, world models, robot policy reward modeling, and reinforcement learning. Previously, I worked on end-to-end drone navigation, legged locomotion, and differentiable simulation.

qchen23 [at] stanford.edu WeChat: CQZ_David
Research direction

Learning robots that improve through experience

I work across policy learning, predictive models, and autonomous systems to make robots more capable, adaptable, and useful in the physical world.

01 · MANIPULATION

Robot manipulation

Building versatile robot policies for complex, long-horizon manipulation tasks in homes and factories.

VLA models Long-horizon tasks Loco-manipulation
02 · EXPERIENCE

Learning from experience

Enabling robots to improve from demonstrations, autonomous rollouts, reward models, and reinforcement learning.

Reward models Self-improvement Reinforcement learning
03 · WORLD

World models

Learning predictive representations of visual appearance, geometry, and dynamics for planning, control, and policy training.

3D representations Differentiable simulation Model-based control
Selected work

Publications

View full publication list →
Preprint

SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation

Q. Chen, H. Zheng, J. Yu, S. Huang, J. Sun, K. Goldberg, C. Wen, P. Abbeel, Y. Shentu, P. Wu, M. Schwager

A multi-task, stage-aware reward modeling framework that produces dense rewards for long-horizon manipulation, enabling VLA policies to improve from low-cost autonomous rollouts.

ICLR 2026

SARM: Stage-Aware Reward Modeling for Long Horizon Robot Manipulation

Q. Chen, J. Yu, M. Schwager, P. Abbeel, F. Shentu, P. Wu

A video-based reward modeling framework that derives progress signals from natural-language stage annotations for scalable long-horizon imitation learning.

CoRL 2025

ParticleFormer: A 3D Point Cloud World Model for Multi-Object, Multi-Material Robotic Manipulation

S. Huang, Q. Chen, X. Zhang, J. Sun, M. Schwager

A 3D world model trained directly from point clouds for dynamics prediction and model-based visuomotor control across objects and materials.

IEEE RA-L 2025 Best Paper Award · 5/1700

GRaD-Nav++: Vision-Language Model Enabled Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamics

Q. Chen, N. Gao, S. Huang, J. Low, T. Chen, J. Sun, M. Schwager

A lightweight onboard vision-language-action framework that trains in a 3D Gaussian Splatting simulator and follows natural-language drone commands in real time.

Updates

News & recognition

Started at Amazon Fauna as an Applied Scientist Intern.

GRaD-Nav++ received the IEEE RA-L 2025 Best Paper Award.

SARM was accepted to ICLR 2026 and added to Hugging Face LeRobot.

GRaD-Nav++ was accepted to IEEE Robotics and Automation Letters.

Admitted to the Stanford Aeronautics & Astronautics PhD program, advised by Mac Schwager.

Education

  • PhD, Aeronautics & Astronautics, Stanford University
  • MS, Mechanical Engineering, Stanford University, 2025
  • BS/BEng, Mechanical Engineering, UIUC & Zhejiang University, 2023

Honors

  • IEEE RA-L Best Paper Award, 2025
  • Stanford Aero-Astro PhD Fellowship, 2025
  • Outstanding Undergraduate Thesis Award, ZJU, 2023
  • First Class Academic Scholarship, ZJU-UIUC Institute, 2022

Academic service

  • Reviewer: IEEE TRO, IEEE TRL, IEEE RA-L, IEEE IoT, IEEE TIE
  • Reviewer: CoRL 2026, NeurIPS 2026, IROS 2025 2026, ICRA 2026
  • Member, IEEE Robotics and Automation Society