Field-tested insights for humanoid teams
Guides, spec sheets, technical articles and benchmarks. Everything we have learned from running real human-movement capture inside humanoid training programs.
Build your own humanoid teleoperation pipeline
A technical case study of the Xsens Humanoid Teleoperation Demo Pipeline: how live Xsens motion capture becomes real-time joint commands on a Unitree G1, and how to assemble the same pipeline yourself from open-source parts. Written for the engineer who has to build it.
✓ Four ways to get suit data in
✓ Retargeting compared
✓ The dual-rate control loop
✓ The safety state machine and engage ramps
✓ Appendix of open-source building blocks for every stage
Xsens Guide: Teleoperation pipeline creation
Humanoid resources
Pick the one that fits where you are right now.
Hardware spec sheet
Full hardware options, latency budgets, range, battery, ARM64 deployment notes.
ROS2 integration guide
Technical handbook for piping Xsens Humanoids motion data into ROS2. Publishers, topics, message types, sample launch files and onboard ARM64 deployment notes.
Download sample data
A free sample of Xsens Humanoids motion files, prepared for robot retargeting: pick-up and placement, bending, floor-to-stand recovery.
Xsens webinars and demos for humanoid teams
Practitioner-led sessions on motion capture for humanoid training, retargeting, teleoperation and human-machine interaction.
Training the humanoid robots of the future
Gijs Heutink’s industry pitch at the ICRA anniversary event. How Xsens motion capture, teleoperation and humanoid shadowing combine to give humanoid teams the high-fidelity human movement data they need.
Discover Xsens Link for humanoid motion training
A walkthrough of the new Xsens Link suit and Xsens Humanoid Software: how clean, robot-ready kinematics get from the operator to the humanoid for teleoperation, dataset creation and simulation seeding.
Humanoid motion training at scale
How humanoid teams move from one operator and one suit to repeatable, multi-operator capture programs. Scalable motion data that does not compromise accuracy, framed for AI training pipelines.
Human-Robot Interaction (HMI / HRI)
From the archive (2015): Dr. Raffaella Carloni (University of Twente) on early human-robot interaction research using Xsens motion capture. The hardware and software have moved on, but the underlying problems she frames are still the ones humanoid teams are solving today.
Insights in your inbox
We email you when there’s something worth reading.
Unsubscribe any time.