German Researchers Create Artificial Nervous System for Robots to Experience Pain
Researchers from Leibniz University of Hannover are currently developing an artificial nervous system for robots to teach them how to feel and react to pain.
At present, robots are being sent to dangerous environments to perform certain actions that are considered to be unpleasant or even deadly for humans due to their inability to feel pain. However, a pair of German researchers believes that letting robots experience pain and teaching them how to properly respond to the stimuli can provide better and safer robot performance.
"Pain is a system that protects us," Johannes Kuehn, one of the researchers, told Spectrum IEEE. "When we evade from the source of pain, it helps us not get hurt."
For the project, presented last week at the IEEE International Conference on Robotics and Automation (ICRA) in Stockholm, Sweden, researchers used "nervous robot-tissue model that is inspired by the human skin structure" to determine how much pain the robot should feel for a given amount of force.
The model works just like human neurons, transmitting pain information in repetitive strikes if the force goes beyond certain thresholds. The researchers classify pain into light, moderate and severe, with robot reacting differently for each classification.
In order to demonstrate the robot's reaction to different classification of pain, the researchers mounted a BioTac tactile fingertip sensor, which can sense pressure and temperature, on a Kuka arm.
For light pain, the robot will experience mild discomfort and will withdraw from the stimuli until the sensation is over. For moderate pain, the robot will withdraw quicker and farther from the stimuli. For severe pain, the robot will switch to passive mode, to avoid further damage.
According to the researchers, letting robots feel and react to different levels of pain may provide safer environment for the robot and human worker that work alongside with it. They noted that undetected damages in robots can lead to accidents causing potential danger for the humans in its proximity.