Researchers at Cornell Engineering developed a soft robot that can recognise when and where it has been injured and repair itself immediately using optical sensors and a composite material. Rob Shepherd, an associate professor of mechanical and aerospace engineering, directed the study. On December 7, 2022, a paper titled “Autonomous Self-Healing Optical Sensors for Damage Intelligent Soft-Bodied Systems” was released.
Shepherd claims that their lab is constantly working to make robots more resilient and agile so they can operate for a longer period of time and with more skills. Unfortunately, if a robot operates for a long time, damage will eventually accrue.
The robot needs to be able to recognise what needs to be addressed as the initial stage in damage restoration. Stretchable fiber-optic sensors have been used for years by Shepherd’s Organic Robotics Lab to create soft robots and associated parts. In these sensors, an optical waveguide transmits light from an LED, and a photodiode detects variations in the intensity of the beam to ascertain when the material is distorted.
Seniors and a polyurethane urea elastomer with hydrogen bonds for quick healing and disulfide exchanges were coupled by the researchers. The SHeaLDS was mounted in a soft robot that resembled a four-legged starfish and was fitted with feedback control by the researchers to demonstrate the technology. The robot recognised the damage when the researchers punctured one of its legs six times, and each wound quickly self-healed. The robot was also capable of changing its gait on its own depending on the damage it detected.
In order to eventually construct “a extremely durable robot that has a self-healing skin, but uses the same skin to sense its environment to be able to do more activities,” Shepherd hopes to link the SHeaLDS with ML algorithms that recognise tactile events.
