Self-Service
Free with hardware purchase
Hardware Cost Only
- Buy hardware from us
- Get the SDK (
rc connect) free - Set up your own teleoperation and recording
- Best for teams with machine-learning experience
Teleoperation Platform
From hardware to trained policy. Control any robot, record demonstrations and train AI models—all from one platform.
17+ Robots4 Input MethodsReal-time ControlAutomatic Data Recording
Three Tiers—Pick the Level That Fits
Choose raw hardware, a recording-ready kit or a fully managed data-to-policy pipeline.
DataKit
Recording-ready out of the box
$2,500 one-time setup
- Everything in Self-Service, plus:
- Camera synchronization configured
- Multi-stream recording ready
- Quality scoring enabled
- LeRobot / OpenPI
- 10 sample episodes included
Best for teams that want to start collecting data immediately
Get DataKitFull Platform
End-to-end data-to-policy
$5K–15K setup
+ $2–3K/month
- Everything in DataKit, plus:
- Cloud data upload
- Automatic quality scoring on every episode
- ACT / Diffusion Policy / VLA
- Deploy back to the robot
- Dedicated support
Best for teams building production robot AI
Contact UsSupported Hardware
Use our broad range of robots, hands and input devices. Buy or rent from us, or bring your own.
Robot Arms
6-DOF — from $3,000
7-DOF — from $8,000
8-DOF — from $4,500 (we manufacture)
Bimanual — from $35,000 (we manufacture)
from $8,500
Industrial collaborative
Adaptive force control
Cost-effective 6/7-DOF
Dexterous Hands
20-DOF, tactile sensing
Dexterous six-finger hand
Research-grade manipulation
Anthropomorphic design
Compact six-finger hand
Input Devices
Full 6-DOF hand tracking
Electronic-skin feedback
Haptic feedback
Basic control
Humanoids
Compact humanoid
Full-size humanoid
Research humanoid
Task-focused humanoid
See the full hardware catalog for specifications and availability, or explore Wuji Glove for dexterous-hand teleoperation. Leasing is available on all platforms.
How It Works
Four steps from zero to a trained policy running on your robot.
1
Buy or Rent Hardware
Choose a robot arm, dexterous hand or humanoid from our store. Hardware ships in days, and leasing is available for teams that want to try before committing.
2
Install the SDK
pip install centeros-cli && centeros connect
One command connects your robot, cameras and input device and automatically discovers hardware on the network.
3
Teleoperate and Record Demonstrations
Control the robot with VR, gloves, a gamepad or a leader arm. Every session is automatically recorded with synchronized video, joint states and actions.
4
Train a Model and Deploy It Back to the Robot
Upload your data, select a training recipe (ACT, Diffusion Policy or VLA) and receive a trained policy. Deploy it to the same robot with one click.
Data Pipeline
Raw teleoperation recordings become deployment-ready policies through a six-stage pipeline.
RecordQuality ScoreCleanExportTrainDeploy
Export Formats
Export MCAP, HDF5, LeRobot (Parquet) and OpenPI in the format your training pipeline needs, without conversion scripts.
Automatic Quality Scoring
Automatically detect stalls, failures, resets and out-of-distribution episodes. Every recording receives a quality score so you train only on clean data.
37%usable dataOnly 37% of Raw Teleoperation Data Is SFT-Usable
False starts, operator hesitation and recovery motions can harm policy training. Our scorer identifies the useful demonstrations so you do not waste compute on noise.
Training Models
Upload your data, choose an architecture and receive a trained policy. We support the models that matter for robot manipulation.
ACT
Action Chunking Transformer. Fast to train and effective with 50–200 demonstrations for single-arm tasks; the default for many teams starting out.
Diffusion Policy
Denoising diffusion for action generation. Handles multimodal action distributions and works well for contact-rich and bimanual tasks.
Tactile-VLA
A Vision-Language-Action model with tactile input for language-conditioned, multi-task policies using force and touch sensing.
Pi0 / Spirit
Fine-tune pre-trained open-source foundation models on your data for faster convergence with less data.
Upload your data and receive a trained policy. See our AI Models page for architecture details and benchmarks.
Ready to Start?
Tell us about your robot, task and timeline. We will scope the right tier and get you moving.
Email us directly: contact@roboticscenter.ai