The Surging Bond Between Additive Manufacturing and the Medical Field
The merger of additive manufacturing and the medical field has carved a path of significant growth. This alliance has given rise to a multitude of applications, spanning from the creation of 3D printed blood vessels and implants to the production of medication in pill form. These groundbreaking advancements are reshaping the landscape of the medical sector. Notably, ETH Zurich has been at the forefront of this journey, consistently introducing remarkable innovations.
In previous endeavors, researchers from ETH Zurich astounded the world by successfully creating 3D printed skin using fungi and developing insoles with real-time monitoring capabilities. Their latest breakthrough resides in the realm of robotics – the introduction of the groundbreaking “Faive Hand.” This tendon-controlled robotic hand, created via 3D printing, is the result of a collaborative effort with the Max Planck ETH Center for Learning Systems. The primary objective was to develop a robotic hand that closely mirrors human attributes.
One glance at the Faive Hand reveals its striking resemblance to a human hand. This lifelike quality empowers it to undertake everyday tasks, including household chores. Its fluid and unrestricted motions allow for precise execution of intricate and skillful activities. Unlike traditional control approaches based on models, the researchers opted for reinforcement learning (RL) as the driving force behind the Faive Hand. This decision was influenced by RL’s suitability for lifelike and intricate robotic configurations that require subtle and delicate actions and attributes.
The Faive Hand, originating from the Soft Robotics Lab, is fabricated using 3D printing technology and propelled by servo motors. The incorporation of rolling contact joints imbues it with the capability for rotations without a fixed axis. However, the intricate design with a high Degree of Freedom (DoF) introduces challenges in terms of management and maneuverability. To address this, joint angle encoders are being developed to enhance accuracy and control during steering operations.
The hand’s simulation framework and low-level controller seamlessly integrate reinforcement learning (RL) training, enabling the execution of commands. This showcases the successful transfer of skills and capabilities within the IsaacGym simulator. Researchers are diligently working to improve the Faive Hand’s capabilities and performance, thereby opening up new possibilities in the realm of human-machine interaction (MMI) in the future.
To stay updated with the latest developments, please visit Faive’s official website. We would love to hear your thoughts on the biomimetic Faive Hand. Leave a comment below or reach out to us via LinkedIn, Facebook, or Twitter. Don’t miss out on our free weekly Newsletter that delivers the latest 3D printing news straight to your inbox. You can also find all our videos on our YouTube channel.
Cover photo credit: ETH Zürich
