Engineers at Cornell University have developed a soft robot that can identify when and where it has been wounded and then mend itself on the spot.
“Our lab is continuously attempting to make robots more durable and agile so that they can function for longer periods of time and with greater capabilities,” said Rob Shepherd, associate professor of mechanical and aerospace engineering. “When you put robots to work over an extended period of time, they will sustain harm. So, how can we let them fix or cope with the damage?”
The Organic Robotics Lab at Shepherd University has created stretchy fibre-optic sensors for use in soft robots and associated components ranging from skin to wearable electronics. They publish their findings in Science Advances.
Shepherd believes that the important first step towards self-healing is for the robot to recognise that there is, in fact, something that needs to be addressed.
Researchers pioneered a technology that uses fibre-optic sensors paired with LED lights to detect minute changes on the surface of the robot. These sensors are coupled with a polyurethane urea elastomer that combines hydrogen bonding and disulfide exchanges for strength and quick healing.
The resultant SHeaLDS (self-healing light guides for dynamic sensing) delivers a damage-resistant soft robot that can self-heal from wounds at room temperature without the need for external intervention.
The researchers implanted the SHeaLDS in a soft robot resembling a four-legged starfish and outfitted it with feedback control to showcase the technology. After researchers pierced one of its legs six times, the robot was able to identify the damage and self-heal each incision in less than a minute. The robot could also change its stride independently depending on the damage it detected.
The material is strong but not indestructible.
“They have qualities akin to human flesh,” Shepherd said. “Burning, acid, or heat do not promote healing because they alter the chemical makeup of the body. However, we are capable of recuperating from wounds.”
Shepherd intends to combine SHeaLDS with machine learning algorithms capable of identifying tactile events in order to build “a very lasting robot with a self-healing skin but utilises the same skin to sense its surroundings in order to execute additional jobs.”
You might also be interested in reading about An architecture that provides consumers with complete control over their devices.
