Accelerate humanoid robot training with real-time motion capture

Accurate, real-time kinematics for training, control, and performance optimization — trusted by global robotics innovators

 

 

Request pricing

Why choose Xsens for humanoid robots training?

Unmatched motion data precision

Scientifically validated motion data for streamlined AI/ML training

Easy integration into your pipeline

- ROS, Unity, Unreal, etc. compatibility
- SDK available
- Full technical support

Scale without limit


- No additional setup needed
- System ready in 15 minutes
- Reliable data even in metal-heavy environments

Request pricing

How does robot training with Xsens Motion Capture work?

1

A human operator wearing an Xsens suit controls the robot's movements in real time, capturing precise motion data

2

The robot collects sensor data during teleoperation to enhance its understanding of movement and environment

3

Combined motion and sensor data are used for AI deep learning, refining the robot's movement patterns

4

The robot autonomously performs learned movements in various conditions using AI-driven insights

All you need for robot training

Get a plug-and-play solution for your pipeline with the three products tailored for humanoid robots training, offering full-body motion capture and sophisticated data processing:

Body movement capture:
Xsens Link

17 embedded sensors (plus 1 extra sensor)

Full body Lycra suit available in 6 sizes (S-4XL)

Update rate up to 240 Hz

150 m range

10 hour battery life

Learn more

Hand movement capture:
Xsens Metagloves

Finger caps, finger tapes, or full liner gloves (S, M, L)

Wireless range up to 15 meters/50 feet

Absolute position, 3 axis rotation

Sensor sample rate 120 Hz

3 hours battery life (swappable batteries)

Data collection and processing:
Xsens Analyze

Real-time streaming (positions, orientations, full-body kinematics)

High data quality (Live Engine + HD reprocessing)

Full magnetic immunity

Flexible exports (MVNX, C3D, BVH, etc.)

Multiple subject support

Get in on the revolution in human-robot interaction