Researchers at Cornell Engineering combined optical sensors with a composite material to create a soft robot that can detect when and where it was damaged and heal itself on the spot. The research was led by Rob Shepherd, associate professor of mechanical and aerospace engineering. The paper “Autonomous Self-Healing Optical Sensors for Damage Intelligent Soft-Bodied Systems” was published on 7th December 2022. 

According to Shepherd, their lab is always trying to make robots more enduring and agile so they operate longer with more capabilities. Unfortunately, robots will accumulate damage if they operate for a long time.  

The first step for repairing damage is that the robot must be able to identify something that needs to be fixed. Shepherd’s Organic Robotics Lab has used stretchable fibre-optic sensors to make soft robots and related components for years. In these sensors, light from a LED is sent through an optical waveguide, and a photodiode detects changes in the beam’s intensity to determine when the material is deformed. 

The researchers combined the seniors with a polyurethane urea elastomer incorporating hydrogen bonds for rapid healing and disulfide exchanges. To demonstrate the technology, the researchers installed the SHeaLDS in a soft robot resembling a four-legged starfish equipped with feedback control. After the researchers punctured one of its legs six times, the robot detected the damage and self-heal each cut in about a minute. The robot could also autonomously adapt its gait based on the damage it sensed. 

Shepherd plans to integrate the SHeaLDS with ML algorithms that recognise tactile events to eventually create “a very enduring robot that has a self-healing skin, but uses the same skin to feel its environment to be able to do more tasks.”