Meet the SharpaWave: an advanced human-sized robotic hand with 22 active degrees of freedom that can take on real-world tasks with human-level dexterity. It relies on sensors and a neural network-based algorithm for advanced tactile sensing. This robot hand can open and close 4 times per second. It is ROS 2 compatible and ready for Python and C++ programming.
The SharpaWave has an open source software stack. It is compatible with Isaac Gym, Isaac Lab, PyBullet, and MuJoCo simulation platforms. It has safety features, such as compliance control to prevent accidental damage. According to the company, each fingertip has a miniature camera and over 1,000 tactile pixels to capture everything. It relies on 6D force sensing for adaptive grip control and slip prevention.
[HT]
