SmartHand

Smarthand
Smarthand greift einen Apfel
DownloadDownload

The SmartHand was developed as part of the Bavarian-funded SmartHand project to enable brownfield automation. Based on CLASH’s variable stiffness actuation – comparable to the David Hand – simulations using PyBullet and optimised cable coupling were used to develop a robotic hand with a manipulation workspace six times larger than that of the David Hand. The hand was presented to the public for the first time at the German Robotics Conference in 2025.

Technical data

Dimensions

205 mm × 105 mm × 85 mm

Weight

1000 g

Degrees of freedom

12

Payload

40 N thumb, 25 N finger

Power supply

8–26 V

Velocity

900°/s

Communication

Ethercat, RS485

Features

  • The modular finger design allows for different kinematics
  • integrated control
  • Self contained design allows it to be used on any cobot

System description

The focus of the development was to design a robust gripper with an improved workspace for in-hand manipulation. To this end, various kinematic configurations were tested in simulation with regard to gripping capacity and in-hand manipulation. The solutions identified were then built and optimised for maintainability and operational robustness. A further objective was to achieve good scalability of the system. To this end, standard-sized model-building servos were installed. These can be fitted with various characteristics, such as high torque, high deceleration or integrated torque control. This allows the user to tailor the hand’s characteristics to their specific use case.

Smarthand
Smarthand greift einen Apfel
Image: 1/4, Credit:
DownloadDownload
smarthand_on_rollin_justin.jpeg
Smarthand montiert auf Rollin Justin.
Image: 2/4, Credit:
DownloadDownload
new_kinematic.jpeg
Die nicht anthrophmorphe Kinematik in der Smarthand erlaubt es, Powergrasp in unterschiedlichen Position zu erreichen. Dies war z.B. mit der DLR Hand II nicht möglich.
Image: 4/4, Credit:
DownloadDownload

Selected publications

Related topics

SmartHand (project)

The aim of the project is the development of dexterous, robust and cost-efficient multi-fingered robotic hands in combination with control based on modern artificial intelligence methods (especially machine learning) for use in industrial automation.