ABB Robotics and PSYONIC Leverage Human-Generated Data to Enhance Robot Dexterity

ABB Robotics is partnering with California-based bionics company PSYONIC to develop robotic grasping and dexterity capabilities using a novel approach that leverages real-world manipulation data from human prosthetics use. By combining the PSYONIC Ability Hand with ABB’s GoFa cobot, the collaboration will investigate how touch and motion data generated from prosthetic use can be utilised to train robots to perform complex and variable tasks that have traditionally been difficult to automate.

“Human dexterity and the instinctive understanding of how to handle different objects is one of the hardest things to replicate in industrial robotics, yet it is a fundamental requirement for truly autonomous and versatile robots,” said Marc Segura, President of ABB Robotics. “As we develop the next generation of physical AI, robots will learn and understand the world as we do. Working with PSYONIC will help close the long-standing gap between human and robotic dexterity, opening new possibilities across industries.”

Grasping and dexterity are central to Autonomous Versatile Robotics (AVR), ABB Robotics’ vision for robots that can sense, think, move, and handle objects with precision in dynamic environments. These capabilities are also crucial for advancing physical AI in industry—robotic systems that learn from real-world interactions and apply that intelligence with industrial-grade reliability. The collaboration will explore new applications across sectors including automotive, aerospace, packaging and logistics, and life sciences. By creating robots that can perform repetitive, ergonomically challenging, or difficult-to-scale tasks, ABB Robotics and PSYONIC aim to help humans and robots work together more effectively while improving productivity, flexibility, and workplace safety.

PSYONIC is working closely with ABB Robotics’ R&D team on integration and development to explore how touch-based manipulation can support next-generation autonomous robotics applications. The PSYONIC Ability Hand was originally developed for prosthetics, combining myoelectric control, tactile sensors, and flexible mechanics in a lightweight, multi-jointed design. Its pressure sensors and vibrotactile feedback system enable users to detect contact, grip force, and release, while the compliant fingers naturally adapt to irregular and deformable objects.

“Dexterous manipulation is fundamentally a data challenge as much as a hardware one,” said Dr. Aadeel Akhtar, Founder and CEO of PSYONIC. “By using the same Ability Hand on both humans and robots, we can gather high-fidelity real-world data on motion, contact, and grip force, then use that data to train robotic systems more effectively. Integrating with ABB Robotics’ platform allows us to scale across more diverse environments and unlock the level of dexterity needed to solve automation’s toughest challenges.”

To support this effort, ABB’s GoFa cobot provides the accuracy and consistency required for industrial-scale deployment, ensuring that subtle variations in grip force, finger position, and motion can be executed and evaluated reliably. This level of precision is essential for translating human-derived manipulation data into dependable robot performance across complex and varied tasks. The collaboration will assess how the combined capabilities can be applied to industrial use cases where conventional gripping technology struggles with variability, fragility, or complexity, such as handling irregularly shaped or delicate items. According to the International Federation of Robotics, advanced gripping technologies and digital integration can reduce development time by up to 30%, underscoring the importance of end-of-arm tooling in accelerating implementation and improving return on investment in automation.

This collaboration also reflects ABB Robotics’ broader strategic approach of working with partners across its ecosystem to overcome persistent automation barriers. By combining robotics, AI, and real-world manipulation data generated through prosthetic use, ABB Robotics continues to advance physical AI and create more sophisticated, adaptive robots capable of operating reliably in real-world environments.