The BrainCo Revo 2 Touch Right (XTR) is a bionic dexterous robotic hand for the right side of a humanoid robot or for a single robot arm, with a capacitive tactile sensor in every fingertip. It packs 11 degrees of freedom, six of them motor-driven, into 383 g, 160 mm from palm base to middle fingertip and 78.5 mm across the palm.
Each fingertip sensor collects six channels and reports pressure, friction, force direction and proximity: normal and tangential force up to 25 N at 0.1 N resolution, and an approaching object up to 1 cm away, sampled at 70 Hz. In tactile adaptive control the hand sets finger current from those readings, so it closes firmly on a bottle and gently on a paper cup. BrainCo's open Python grasping example detects contact and estimates object stiffness from the same data, pinching a foam block with two fingers, holding a paper cup with three and gripping a water bottle with five, a ready starting point for grasp learning, teleoperation and contact-rich manipulation research.
Under the sensors the Touch is the same mechanism as the rest of the Revo 2 line: at least 50 N grip, 15 N pinch, a 20 kg payload, 0.1° repeat precision and a full flex in 0.65 s or less. The Revo 2 Basic right hand (XRR) runs RS-485 or CAN FD on 12–28 V, the Revo 2 Pro right hand (XER) adds EtherCAT and a 12–64 V input, and the Touch keeps the Pro's interfaces and adds the fingertip sensing.
Its default device ID is 127, so it shares a bus with the Revo 2 Touch left hand (XTL), which is set to 126, without re-addressing. Control runs at up to 1 kHz over EtherCAT and 500 Hz over RS-485 or CAN FD, from BrainCo's Python and C/C++ SDKs on Linux, Windows and ROS, with URDF and MuJoCo models for simulation. The box holds the hand with its RS-485/CAN FD, EtherCAT and power cables. For a Unitree H2 there is a Revo 2 tactile hand configured for the H2, and the rest of our dexterous robotic hands cover other makers and designs.
Key Capabilities
-
Fingertips that feel: Each of the five fingertips carries a capacitive sensor that reads pressure, friction, force direction and proximity, resolving 0.1 N up to 25 N.
-
Grip set by touch: Tactile adaptive control sets finger current from the fingertip readings, so a paper cup and a water bottle each get the force they need.
-
Senses before contact: The proximity channel notices an object up to 1 cm from a fingertip, and the tactile data is sampled at 70 Hz.
-
383 g, 20 kg payload: The right hand weighs 383 g yet carries at least 20 kg, with at least 50 N five-finger grip and 15 N pinch.
-
1 kHz over EtherCAT: EtherCAT with CoE and PDO runs at up to 1 kHz, RS-485 and CAN FD at up to 500 Hz, on any supply from 12 to 64 V.
-
Ready for ROS and simulation: Python and C/C++ SDKs run on Linux, Windows and ROS, with URDF and MuJoCo models and over-the-air firmware updates.
Buying through Robotix Market
This unit is sourced new through established distribution channels and ships worldwide with the manufacturer warranty. Technical support is coordinated with the manufacturer's own service network. Deployment scope, configuration, and commissioning responsibilities are agreed as part of the quotation — on-site installation is not included in the product price.