Network

Physical-world capture at scale

The network is the field operation behind ROBOTRAIN: the sites, rigs, protocols, and people that turn real behaviour into training data. Nothing here is synthetic.

Field workflow

How a session runs

A predictable sequence from scouting to upload — so quality stays high as we scale locations and operators.

  1. 01

    Site selection

    We choose capture sites for task diversity — kitchens, workshops, warehouses, and offices. Sites are scouted for lighting, clearance, and mobility constraints that mirror real robot deployments.

  2. 02

    Rig deployment

    Operators deploy standardised kits: head and wrist POV (RGB + depth), IMU logging, and optional controller telemetry. Rigs are calibrated before day one and rechecked between sessions.

  3. 03

    Capture protocols

    Structured task scripts — pick and place, handover, navigation under occlusion, bimanual work — make coverage intentional. Deviations are logged; scripts are versioned with each release.

  4. 04

    Sync & handoff

    Session bundles are encrypted and uploaded securely. The ROBOTRAIN platform runs intake checks and queues annotation. Operators receive a quality summary the next morning.

Quality & governance

Standards you can cite

Researchers and enterprises need data they can publish and ship. Our field and data standards are built for that bar.

  • Informed consent before every session
  • Faces and identifiers blurred before data leaves the capture site
  • Chain-of-custody from device to release
  • Scripts and calibration logs retained with each dataset version
  • Raw footage access limited to named ROBOTRAIN engineers

Partners

Join the capture network

Property owners & operators

We work with managers and homeowners for structured on-site sessions. Compensation and consent are agreed upfront. All releases are anonymised before they leave our control.

Hardware manufacturers

Integrate your sensors into our rig standard. Approved partners receive pre-release samples for benchmarking against the public corpus.

Annotation partners

We work with teams experienced in robotics labelling. If you have capacity for spatial and temporal tasks, we would like to hear from you.

Interested in partnering?

We are building partnerships across sites, hardware, and annotation. More on how the network operates is on this page; formal partner onboarding will be announced later.