English 中文 日本語 Русский
NOKOV Showcases Banner

Capturing Motion,
Crafting Stories

Explore Our Case Studies: Transforming Motion into Masterpieces Across Industries

How Motion Capture Enables MuJoCo Simulation and Teleoperation for the Unitree G1 Humanoid Robot

Client
Ningbo Polytechnic University
Capture volume
Application
Humanoid robot, MuJoCo, Unitree G1, teleoperation, human motion tracking
Objects
Human body
Equipment used

Overview

Ningbo Polytechnic University uses the NOKOV Motion Capture System together with the Unitree G1 humanoid robot to perform MuJoCo simulation training and real-time teleoperation.

During the experiment, an operator wearing a motion capture suit with reflective markers performs upper-body movements such as arm lifting, arm extension, and waving. The NOKOV motion capture system captures human motion in real time and streams high-precision pose data to both the MuJoCo simulation environment and the physical robot, enabling synchronized motion retargeting and teleoperation.

Experimental Workflow

The experiment consists of two stages.

Human Motion Capture

The operator wears a motion capture suit equipped with reflective markers and performs natural upper-body movements inside the capture area.

The NOKOV Motion Capture System continuously tracks the operator's motion and outputs accurate, low-latency pose data for robot control.

MuJoCo Simulation

Captured motion data is streamed directly into the MuJoCo simulation environment.

The Unitree G1 robot model reproduces the operator's movements in real time, allowing researchers to validate robot behavior before deployment to the physical platform.

Real-Time Robot Teleoperation

Following simulation, the same motion data drives the physical Unitree G1 humanoid robot.

The robot synchronously performs arm lifting, waving, and other movements, demonstrating real-time interaction between human motion input and robot execution.


Application Value


FAQ

Q1:Why use motion capture for humanoid robot teleoperation?

A1:Motion capture provides accurate and low-latency human pose data that can be mapped directly to a humanoid robot, enabling intuitive teleoperation and reliable robot control.

Q2:How does motion capture work with MuJoCo?

A2:The NOKOV Motion Capture System streams real-time motion data into the MuJoCo simulation environment, allowing the virtual robot to reproduce human movements for algorithm development and validation.

Q3:Can the same motion capture data drive both simulation and a real robot?

A3:Yes. In this application, the same human motion data is used to control both the Unitree G1 model in MuJoCo and the physical Unitree G1 humanoid robot, creating a seamless workflow from simulation to real-world teleoperation.


Learn more about the application of NOKOV mocap on Humanoid Robotics&Embodied AI

Related Products

NOKOV Optical Motion Capture System Mars Series

NOKOV ShadowEngine Embodied AI Training Platform

Prev
IEEE RA-L | GeoPF: Infusing Geometry into Potential Fields for Reactive Planning in Non-trivial Environments

NOKOV Motion Capture Basketball Game Demo

UMI Game
2022-03-29

Kung Fu Motion Capture Performance

Shu-Gu Entertainment
2023-02-06

Applications of motion capture systems in wire-driven continuum robot research

Sichuan University
2022-06-17

RA-L Paper: Enhancing Generalizable 6D Pose Tracking of an In-Hand Object With Tactile Sensing

Tsinghua University
2024-12-12

By using this site, you agree to our terms, which outline our use of cookies. CLOSE ×

AI Chatbot
Hello! I'm the AI assistant of NOKOV. How may I help you today?
Contact us
We are committed to responding promptly and will connect with you through our local distributors for further assistance.
Engineering Virtual Reality Life Sciences Entertainment
I would like to receive a quote
Beijing NOKOV Science & Technology Co., Ltd (Headquarter)
LocationRoom820, China Minmetals Tower, Chaoyang Dist., Beijing
Emailinfo@nokov.cn
Phone+ 86-10-64922321
Capture Volume*
Objective*
Full Bodies Drones/Robots Others
Quantity
Camera Type
Pluto1.3C Mars1.3H Mars2H Mars4H Underwater Others/I do not know
Camera Count
4 6 8 12 16 20 24 Others/I don't know