Insider Brief
- ABB Robotics and Psyonic are collaborating to explore whether touch, grip and motion data generated by people using robotic prosthetic hands can be used to train robots to perform complex manipulation tasks with greater dexterity.
- The partnership integrates Psyonic’s Ability Hand prosthetic with ABB’s GoFa collaborative robot to study how human-generated manipulation data can improve robotic handling of delicate, irregular and variable objects across industries including automotive, aerospace, logistics, packaging and life sciences.
- ABB said the work supports its vision for more adaptable physical AI systems, with the companies investigating how real-world human interaction data can help robots learn tasks that have traditionally been difficult to automate reliably.
ABB Robotics is partnering with San Diego-based bionics company Psyonic to explore whether data generated by people using robotic prosthetic hands can help train robots to handle objects with greater dexterity.
The collaboration will study how touch, grip and motion data collected during everyday human use of prosthetic devices can be applied to robotic systems performing tasks that have traditionally been difficult to automate. Abb said the companies plan to explore how that can then be put to use across industries, including automotive, aerospace, logistics, packaging and life sciences.
Under the partnership, Psyonic is working with ABB’s research and development team to integrate the Ability Hand with ABB’s robotics platform. ABB’s GoFa cobot will provide the precision and repeatability needed to evaluate how human-generated manipulation data can be translated into robotic actions.
“Human dexterity and the instinctive understanding of how to handle different objects is one of the most difficult things to replicate in industrial-grade robotics, but it’s a fundamental need for truly autonomous and versatile robots,” said ABB Robotics president Marc Segura. “As we develop the next generation physical AI, robots will learn and understand the world as we do. This collaboration with Psyonic will help to close the long-standing gap between human and robot dexterity, opening up new possibilities for a wide range of industries.”
Psyonic’s Ability Hand was originally developed as a prosthetic device. According to ABB, the system combines touch sensing, pressure detection and multi-jointed fingers that can adapt to objects of different shapes and sizes. Because the same hand can be used by both people and robots, the companies said they see an opportunity to capture real-world data on how humans grasp, move and interact with objects and use that information to improve robotic performance.
“Dexterous manipulation is ultimately a data challenge as much as a hardware challenge,” said Psyonic founder and CEO Dr. Aadeel Akhtar. “By using the same Ability Hand on people and on robots, we can capture high-fidelity real-world data on movement, contact and grip force, then use that to train robotic systems more effectively. Integrating with ABB Robotics’ robotics platform allows us to expand into more environments and unlock the level of dexterity needed to take on the hardest challenges in automation.”
ABB said the work aligns with its work on what it calls “Autonomous Versatile Robotics,” or robots capable of sensing, reasoning, moving and manipulating objects in changing environments. The company suggests dexterous manipulation is a key requirement for future physical AI systems that can learn from real-world interactions and perform tasks with industrial reliability.
According to ABB, the approach could be particularly useful in environments where traditional robotic grippers struggle with fragile, irregular or highly variable objects. By improving a robot’s ability to adjust grip force and finger positioning, the companies hope to expand the range of tasks that can be automated.
