Frequently Asked Questions (FAQ)

1. What is the Dexterous Hand SDK? The Dexterous Hand SDK is a Python library provided by Stonedrum Robotics to control and interface with the L20 dexterous hand.

2. Which Python versions are supported? The SDK requires Python 3.10 or newer (3.10, 3.11, and 3.12 are tested in CI).

3. What operating systems are supported? We officially support Ubuntu Linux and macOS. Windows is supported via WSL2.

4. Which ROS 2 distributions are supported? We currently support ROS 2 Humble (Ubuntu 22.04) and ROS 2 Jazzy (Ubuntu 24.04).

5. Can I use the SDK without ROS 2? Yes, the core dexterous_hand Python package is completely independent of ROS 2 and can be run standalone.

6. Can I simulate the hand without physical hardware? Yes. You can use the built-in Mock Mode for software-in-the-loop testing, or use our provided URDF models in simulators like Gazebo or MuJoCo.

7. How do I enable Mock Mode? Simply initialize the hand using the mock factory method: hand = Hand.mock(). This requires no physical connection.

8. How do I define a custom grasp pose? You can instantiate a GraspPose object with your desired joint angles and pass the joint_positions_rad dictionary directly to hand.move_joints().

9. What are the standardized joint names? The 5 main joints used in the SDK are thumb_flex, index_flex, middle_flex, ring_flex, and little_flex.

10. Is MoveIt 2 supported? MoveIt 2 configuration packages are in development and will be provided in a future release to enable advanced motion planning.

11. Does the hand require manual calibration? Under normal operation, the hand auto-calibrates upon startup. Manual calibration is only required if an actuator is replaced.

12. Why do I see a NotImplementedError in read_glove_frame? Teleoperation via data gloves is planned for a future update. The method signature exists but is not yet implemented.

13. Is the hardware CE certified? The CE certification is currently under verification. For more details, please contact info@stonedrum.co.

14. How do I report a bug or request a feature? Please use the GitHub Issues tracker on our repository to report bugs or submit feature requests.

15. Who do I contact for commercial enquiries? For enterprise support, volume pricing, or general commercial enquiries, please contact Stonedrum Robotics at info@stonedrum.co.

16. Do you offer training data for dexterous manipulation models? Yes, our Training Data Module is currently in development (targeted for Q4 2026). It will allow you to load third-party and partner datasets directly into your ML pipelines. For more details, see our Training Data concepts page. Contact info@stonedrum.co for early access.