How people with prosthetic arms are teaching robots to ‘feel’ like a human
How people with prosthetic arms are teaching robots to ‘feel’ like a human
How people with prosthetic arms are teaching - Researchers are exploring groundbreaking ways to bridge the gap between human and robotic capabilities, with a focus on enhancing the precision and adaptability of machines through the insights of individuals using advanced prosthetic technology. A recent partnership between ABB Robotics and the California-based bionics firm PSYONIC has sparked excitement in the field, aiming to translate the nuanced skills of human dexterity into robotic systems. This initiative could revolutionize automation by enabling machines to perform complex tasks with greater sensitivity and reliability, particularly in environments where consistency and finesse are critical.
A collaborative leap in robotic dexterity
ABB Robotics, a leader in industrial automation, has partnered with PSYONIC to integrate their prosthetic device, the Ability Hand, into the company’s GoFa collaborative robot. This innovation marks a significant step toward developing robots that can mimic human-like movement and tactile awareness. By leveraging real-world data from prosthetic users, the project seeks to improve the robots’ ability to handle delicate objects, such as fragile items or irregularly shaped materials, which are often challenging for automated systems.
The collaboration centers on the ability of the Ability Hand to generate detailed touch and motion data. This data will be used to train the GoFa cobot, allowing it to adapt to varying conditions and perform tasks with a level of sophistication previously seen only in human hands. ABB’s president, Marc Segura, emphasized that the combination of human intuition and robotic precision could redefine the future of automation. “Humans possess an innate understanding of how to manipulate objects, even when faced with unpredictable environments,” he said. “Reproducing this skill in machines requires not just advanced hardware but also the ability to interpret and apply real-world data effectively.”
From prosthetics to robotics: a shared data model
Dr. Aadeel Akhtar, founder and CEO of PSYONIC, highlighted the dual purpose of the Ability Hand. Originally designed as a prosthetic for amputees, the device now serves as a bridge between human and robotic systems. “The same technology that helps people regain independence can also be used to teach robots how to interact with the world,” he explained. “This creates a unique opportunity to gather high-fidelity data on grip force, movement patterns, and tactile feedback, which can then be used to refine robotic capabilities.”
The PSYONIC Ability Hand employs myoelectric control, allowing users to operate it through muscle signals, and incorporates pressure sensors to detect contact and object release. Its flexible fingers are engineered to adapt to deformable or oddly shaped items, offering a level of versatility that conventional robotic grippers lack. By integrating this hand into the GoFa cobot, ABB and PSYONIC are creating a platform where robots can learn from human interactions, potentially reducing the need for extensive programming and increasing their responsiveness in dynamic settings.
This approach is part of ABB Robotics’ broader vision for Autonomous Versatile Robotics (AVR). The company believes that true automation requires machines to sense, reason, and manipulate objects with the same fluidity as humans. “The challenge lies in translating human dexterity into algorithms that can replicate it,” Segura added. “By using prosthetic data, we’re not just improving technical performance—we’re making automation more intuitive and accessible.”
Applications across industries
The potential applications of this technology are vast. In automotive manufacturing, robots equipped with this system could handle delicate components with greater care, reducing the risk of damage during assembly. In aerospace, the ability to grasp irregularly shaped parts might streamline complex tasks like component testing or maintenance. Packaging and logistics could benefit from robots that adapt to varying product sizes, while the life sciences sector may see improved precision in tasks such as medical device handling or surgical simulations.
Moreover, the collaboration addresses the limitations of traditional robotic grippers, which often struggle with tasks requiring variable force or adaptability. For instance, a robot might struggle to gently lift a glass without crushing it, or to adjust its grip when handling a misshapen object. The PSYONIC-ABB partnership aims to overcome these challenges by enabling robots to learn from human interactions, ultimately making them more efficient and versatile in industrial environments.
Training robots through human data
The integration of the Ability Hand into the GoFa cobot represents a shift in how robotic systems are trained. Instead of relying solely on pre-programmed instructions, robots can now absorb real-world data from prosthetic users, allowing them to adapt to new scenarios on the fly. This method is particularly valuable for tasks that require subtle adjustments, such as adjusting grip strength based on the texture or weight of an object.
Segura noted that the ability to translate human-derived data into consistent robotic actions is crucial for advancing physical artificial intelligence. “In industrial settings, robots must perform with precision, even when faced with unpredictable variables,” he said. “The data from prosthetic use provides a realistic framework for training, ensuring that robotic systems can handle a wide range of challenges without human intervention.”
Dr. Akhtar further elaborated on the significance of this collaboration. “By using the same technology on both people and robots, we’re creating a shared model that can be refined continuously,” he said. “This not only improves the performance of robotic systems but also enhances the functionality of prosthetics, creating a symbiotic relationship between human and machine.”
Real-world implications and future possibilities
The success of this project could have far-reaching implications beyond automation. It may also contribute to advancements in assistive technologies, helping to improve the quality of life for individuals with limb differences. By applying the same principles used to train robots, prosthetic users might gain even more precise control over their devices, further blurring the line between human and machine.
Additionally, the partnership opens the door for robots to take on physically demanding work that is currently reliant on human labor. Tasks such as repetitive assembly, intricate packaging, or handling hazardous materials could be automated more efficiently, potentially reducing workplace injuries and increasing productivity. Segura believes this technology will be a cornerstone in the evolution of industrial automation, enabling robots to operate with the same adaptability as humans.
As the project progresses, it is expected to pave the way for more advanced robotic systems that can learn from real-world interactions. This could lead to machines that not only perform tasks with greater precision but also respond to environmental changes in a way that mirrors human intuition. The combination of PSYONIC’s prosthetic expertise and ABB’s robotics platform is seen as a transformative step in the field, with the potential to redefine automation across multiple sectors.
The collaboration also underscores the growing importance of interdisciplinary approaches in technology development. By merging bionics with robotics, the partnership is addressing a long-standing challenge in automation: the ability to replicate human dexterity in machines. This could lead to new innovations in industries ranging from healthcare to manufacturing, where precision and adaptability are essential.
While the project is still in its early stages, experts believe it has the potential to revolutionize how robots are designed and trained. The ability to use human-derived data not only enhances the performance of machines but also offers a more intuitive and user-friendly approach to automation. As the technology evolves, it may become a standard tool for improving the efficiency and effectiveness of robotic systems in dynamic environments.
For now, the partnership is focused on refining the integration of the Ability Hand with the GoFa cobot. The firms are testing the system’s ability to handle variable tasks, with the ultimate goal of creating robots that can perform complex operations with the same level of skill as a human. This represents a significant milestone in the quest for truly autonomous and versatile machines, with the promise of transforming industries through the power of human-inspired robotics.
As the technology continues to develop, it is expected to set new standards for robotic precision and adaptability. The collaboration between ABB and PSYONIC is not just about improving automation—it’s about creating a new paradigm where machines can learn from humans, adapt to their surroundings, and perform tasks with a level of sophistication that was once thought impossible. This could mark the beginning of a new era in robotics, where the line between human and machine becomes increasingly indistinct.
“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.” – Marc Segura, President of ABB Robotics
With this innovation, the future of automation looks brighter than ever. The integration of prosthetic data into robotic systems is a testament to the power of collaboration and the potential for technology to evolve in ways that benefit both humans and machines. As the project moves forward, it is hoped that the resulting advancements will lead to more efficient, flexible, and safe industrial environments, driven by the ingenuity of human-inspired robotics.
ArrowMORE: ‘Starmer’s AI jobseeker chatbot wrote my pet cat a CV and said she’s employable’
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply. Your information will be used in line with our Privacy Policy.