Skip to content
View RRGGZZ's full-sized avatar
馃榾
Working!!!
馃榾
Working!!!
  • Shanghai Jiao Tong University
  • 800 Dongchuan RD. Minhang District, Shanghai, China

Highlights

  • Pro

Block or report RRGGZZ

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don鈥檛 include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user鈥檚 behavior. Learn more about reporting abuse.

Report abuse
RRGGZZ/README.md

Hi锛両'm Guanzhu Ren

Humanoid Robotics | Whole-Body Control | Robot Dynamics

WBC聽聽 MPC聽聽 BFM聽聽 VLA

M.S. Student at Shanghai Jiao Tong University 聽|聽 Enrolled in 2025 聽|聽 Shanghai, China

Profile

I am an M.S. student at Shanghai Jiao Tong University, enrolled in 2025. My research focuses on motion control for humanoid robots, high-bandwidth harmonic disturbance rejection, robotic mechanism design, and dynamics modeling.

I am interested in control architectures that connect rigorous physical modeling with learning-based behavior representations. My goal is to improve the precision, robustness, and whole-body capability of humanoid robotic systems in demanding real-world tasks.

Current Research

01. Closed-Loop Teleoperation

Developing high-precision, drift-free whole-body robot teleoperation methods and systems based on closed-loop control, with an emphasis on long-horizon stability, motion consistency, and accurate human-to-robot motion retargeting.

02. MPC-Guided Behavior Foundation Models

Developing an MPC-guided Behavior Foundation Model (BFM) framework for high-payload humanoid robots. The work combines predictive guidance with learned behavior representations to pursue dynamically feasible whole-body motion, robust balance control, and coordinated manipulation under substantial payloads.

03. Whole-Body VLA Benchmark

Building a whole-body Vision-Language-Action (VLA) benchmark and data acquisition system for synchronized multimodal observations, whole-body trajectories, robot states, control signals, and task-level semantic annotations.

Selected Engineering Work

Robot-to-Robot Retargeting
Motion retargeting from the Unitree G1 humanoid to the PNDbotics Adam Pro, developed on top of the GMR framework.

Parallel Ankle Simulation
Forward and inverse kinematics, dynamics analysis, and MuJoCo simulation for the PNDbotics parallel ankle mechanism.

TeleVision Lab
An Isaac Lab and Isaac Sim migration workspace for immersive humanoid teleoperation based on Open-TeleVision.

PND Robot Models
Simulation and control assets for the PNDbotics Adam Pro, including URDF/MJCF descriptions, meshes, and supporting model resources.

Research Map

Control and Optimization Modeling and Simulation Embodied Robot Systems
Model Predictive Control Kinematics and Dynamics Whole-Body Teleoperation
Whole-Body Control Mechanism Design Motion Retargeting
Disturbance Rejection System Identification Behavior Foundation Models
Trajectory Optimization MuJoCo / Isaac Sim Vision-Language-Action Systems

Tools

PythonC++MATLAB/SimulinkROS 2MuJoCoIsaac SimIsaac Lab


Model the dynamics. Close the loop. Build the system.

Pinned Loading

  1. Realtime-Retargeter-For-Mocap Realtime-Retargeter-For-Mocap Public

    Forked from NVIDIA/soma-retargeter

    SOMA BVH to humanoid robot motion retargeting library built with Newton and NVIDIA Warp

    Python 1

  2. TeleVision_lab TeleVision_lab Public

    Isaac Lab / Isaac Sim migration workspace for Open-TeleVision.

    Python 2

  3. robot_to_robot_retargeting robot_to_robot_retargeting Public

    We implemented retargeting from the Unitree G1 robot to the Pndbotics Adam Pro robot based on the GMR framework.

    Python 1

  4. g1_deploy_sonic g1_deploy_sonic Public

    Based on the sonic strategy, the G1 is further developed to synchronize audio and video, making it suitable for scenarios such as sales guidance and greeting.

    C++ 1

  5. parallel-ankle-sim parallel-ankle-sim Public

    Forward/Inverse kinematics, dynamics, and MuJoCo simulation for the PNDbotics parallel ankle mechanism.

    Python 3

  6. pnd_models pnd_models Public

    This repository contains PNDbotics adam-pro model files for simulation and control, including URDF/MJCF descriptions, mesh assets, and related resources.

    Python 1