Boston Dynamics unveils four-fingered hand for Atlas — Korea Herald
In South Korea, Boston Dynamics, part of Hyundai Motor Group, unveiled a new four-fingered robotic hand for the Atlas humanoid robot. It was developed so the robot can continuously perform complex tasks in industrial environments, Korea Herald reports.
Tactile sensors and 13 degrees of freedom
The new hand has 13 degrees of freedom: four in the thumb and three in each of the other fingers. Tactile pressure-sensitive sensors are installed on the fingertips and palm. They make it possible to detect minor changes in contact force and adjust finger movements and grip force depending on an object's shape, material and contact conditions.
In a demonstration video, Atlas inserted a drill bit into an electric drill, pulled the trigger and drilled a hole in wood. The robot also tightened a small nut by rotating it with its fingers, spun a thin drumstick and simultaneously manipulated two golf balls.
More current news is available on the UA.News Telegram channel Telegram.
Design for manufacturing
The four-fingered design was developed with industrial use, design complexity, energy consumption, cost-effectiveness, durability and maintenance in mind. The hand is approximately the size of a human hand and has modular components and a simplified structure.
The company tested various options, including the number of thumbs and the need for a little finger, using experiments, simulations and 3D-printed mock-ups. To train Atlas, Boston Dynamics used a Sim-to-Real approach: it first trained the robot in virtual environments and then transferred the results to the physical machine. The simulations accounted for equipment characteristics and friction, and also varied motor torque, object shapes and external disturbances.
The company recently opened the Robotics Metaplant Application Center, which replicates the manufacturing environment of Hyundai Motor Group. The center will collect data on Atlas in production and test its training through data collection, training, deployment in manufacturing conditions, feedback and retraining.
