Force-controlled dexterous robotic hand · serves as: Manipulation, Grasping, Sensing, Human Machine Interaction. · compatible with: Humanoid, Manipulator, Research Robot.
Which group Sharpa Wave Dexterous Hand belongs to and how it is built
A subcategory encompassing robotic hands with multiple degrees of freedom, designed to perform complex grasping, regrasping, manipulation, and object interaction tasks. Unlike simple grippers, dexterous hands enable more advanced finger coordination and a broader variety of grasp types.
Force-controlled dexterous robotic hand components are advanced manipulator end-effectors equipped with force, torque, or contact sensing, designed for safe and precise object handling. They are used in humanoid robots, research platforms, and tasks requiring delicate or adaptive grasping.
Five-fingered robotic hands designed for a wide range of grasps and more anthropomorphic manipulation, commonly used in humanoid robots and manipulation research.
Robots in which Sharpa Wave Dexterous Hand is or can be used
Sharpa Wave is a dexterous five-finger robotic hand built at 1:1 human scale (palm-width to hand-length ratio ~0.618), designed for embodied AI research and advanced robotic manipulation. Its 22 active degrees of freedom in an isomorphic layout mirroring the human hand allow it to manipulate the same tools as humans, with a 10 mm minimum grasp diameter and ±1 mm fingertip position repeatability.
Each finger integrates a torque sensor and a Dynamic Tactile Array (DTA) - a proprietary neural-network-driven tactile sensing array capable of detecting objects as light as a butterfly. Maximum active fingertip force is 20 N, control runs at 500 Hz with movement speed above 4 Hz across all gestures. The hand performs an automatic protective clench within 0.10 s and has been validated through 2.5 million press cycles and 4,000 km of fingertip friction testing.
Sharpa Wave is a component of the NVIDIA Isaac GR00T Reference Humanoid Robot (built on the Unitree H2 chassis): a dual-hand setup contributes to the robot's 75 total degrees of freedom. The vendor ships ROS 2 packages, URDF/MJCF simulation assets for MuJoCo, and provides C++/Python support on macOS and Linux hosts.
