The partnership leverages real-world manipulation data from bionic hands to train robots and reduce industrial engineering time.
ABB Robotics has partnered with California-based bionics company Psyonic to develop new grasping and dexterity capabilities in robots. The project combines Psyonic's Ability Hand prosthetic with ABB's GoFa robotic arm. The initiative employs a novel approach that uses real-world manipulation data collected from the use of human prosthetics.
The dexterity required to manipulate objects of varying shapes remains a major challenge for the automation industry. According to data released by ABB Robotics, improvements in handling and grasping capabilities can reduce the engineering time needed to program and adapt robots by up to 30%.
The core strategy of the collaboration involves using human-generated data to train robotic algorithms. By analyzing how prosthetic users interact with different objects in their daily lives, the companies can extract practical movement and force data. This dataset is then applied to refine the precision of robotic grippers and hands in industrial environments.
The integration of Psyonic's bionics technology with ABB's hardware aims to overcome the current limitations of industrial robotics, which frequently require extensive programming efforts to handle complex parts. Transferring human manipulation patterns to machines seeks to make robots more adaptable to irregular tasks without the need for continuous manual adjustments.
Advancements in robotic handling have direct implications for production line efficiency. With more dexterous robots capable of learning from real-world data, the industry can accelerate assembly and material handling processes, reducing operational bottlenecks and technology implementation costs.
The partnership uses real-world manipulation data collected from human prosthetic users of Psyonic's Ability Hand. This human-generated data is applied to train robotic algorithms, refining the precision and adaptability of robotic grippers in industrial environments.
Improving robotic handling and grasping capabilities can reduce the engineering time needed to program and adapt robots by up to 30%. It also allows robots to handle complex parts without continuous manual adjustments, accelerating assembly and reducing operational costs.
The project integrates Psyonic's bionic Ability Hand with ABB's GoFa robotic arm to develop new grasping capabilities based on human movement and force data.