Paolo Falco Rüegg

Bionic Soft Hand

Sensitive robot hand with artifical intelligence

Interaction Engineering Wearables

This was the main project I was involved in at FESTO Bionics. My focus was developing an IMU sensing system for the soft robotic hand and implementing physical interactions such as teleoperation. I conceptualised and developed three iterations of wearable sensor gloves that could determine the current pose of the hand. These early-stage prototypes were later miniaturised and incorporated into the hand. The bulk of the work, however, was building software on top of sensor system.

“Whether grasping, holding or turning, touching, typing or pressing – in everyday life, we use our hands for the most diverse tasks. The human hand is a true miracle of nature. What could be more logical than equipping robots in collaborative working spaces with a gripper that is based on this natural model and can learn through artificial intelligence to solve a wide variety of gripping and turning tasks?”

In terms of software, I implemented a ‘digital twin’ of the physical hand (visualisation), PID position control and teleoperation. The robot hand was running on ROS, so the bulk of the code was written in Python and C++. Unfortunately, I had to return return to university before the project was completed, but left the team with a thorough documentation of my work and code. Other things I contributed to included:

  • Working with TensorFlow to teach the hand movement patterns.
  • PCB design and manufacture using EAGLE
  • Data collection using my sensor system for subsequent reinforcement learning