Sharpa Wave / SharpaWave, L/R
Specifications
Record source date: 2026-09-16
Active / total DOF: 22 / —
DTA fingertip tactile; up to 180 fps; 20 ms latency; configuration-dependent
Weight basis: Hand only; camera and compute excluded.
Availability / status: Current / marketed
Evidence limits: Manufacturer table shows paired 20/12 N fingertip, 150/90 N grip and 40/24 kg payload; extracted labels do not resolve rating/configuration mapping. No single force value assigned.
Price scope: Seller USD 50,000 for one LEFT Wave; tax and freight extra. Revision/accessories need confirmation. The separate RobotLAB USD 78,200 H2 camera/platform package remains in offer history; these are different offers. Manufacturer mass 1.3 kg retained over seller 0.8 kg field.
Software, projects & datasets
What is ready to build with?
ROS 2; MuJoCo and Isaac Sim models; Sharpa Pilot + Wave SDK
Product source ↗Check each link’s model scope, release, files and license. A reference is not a tested integration.
Dexhands has not reproduced a software setup or hardware test for this record. Published support is not a test result.
Curated public references, checked 2026-09-22. Links describe a model or family relationship, not hardware-tested compatibility.
Sharpa Wave SDK
Sharpa · Model-specific
Manufacturer SDK downloads and integration instructions for Sharpa Wave.
Check platform architecture, release and firmware compatibility.
Sharpa Pilot
Sharpa · Model-specific
Device setup, monitoring and firmware-management application.
Follow the manufacturer update procedure and supported release combinations.
Sharpa robot models
Sharpa · Family resource
URDF, USD and XML assets maintained by Sharpa.
Select the Wave model and check the asset-specific license.
Wave hand retargeting
Sharpa · Model-specific
Camera-based hand tracking and retargeting for Sharpa Wave.
Requires Sharpa Wave SDK and the stated capture dependencies.
Sharpa reinforcement learning lab
Sharpa · Family resource
Manufacturer learning examples in Isaac Lab.
Use the provided hardware model and simulator version; this is experimental code.
Learn to audit a model-to-hand interface →