The Story
“Lego Robot with an Organic ‘Brain’ Learns to Navigate a Maze”
https://www.scientificamerican.com/article/lego-robot-with-an-organic-brain-learns-to-navigate-a-maze/
by Saugat Bolakhe
Scientific American, January 28, 2022
The Pitch
Hi [editor name redacted],
Here, with a new pitch. Let me know what you think.
Reporting in the journal Science Advances on December 10, researchers at the Max Planck Institute of Polymer Research have created a human neuron mimicking device – also called an organic neuromorphic electronic device – that not only learns from experience but also adapts and acts accordingly.
Neuromorphic devices are organic soft materials that help perform multiple computing tasks with low energy. For this new experiment, the researchers placed the neuromorphic device-fit robot in a maze. In navigating the maze, each time the robot made a wrong decision, it received a corrective stimulus in the form of an electrical signal. However, interestingly, with each subsequent attempt, it also went on adapting and avoiding maze traps. Finally, at the 16th attempt, the robot made it to the final exit.
Researchers are intrigued by how a simple organic circuit was able to learn and adapt so swiftly. In the future, they believe, its application could help in advancing brain-machine interface, instant disease diagnostic (Point-of-care diagnostic), and improvising real-world robotics.
Would it be a good fit for short coverage?
(Editor follow up)
Thanks for this! Can you provide more detail about how this particular device would help enable these applications (“advancing brain-machine interface, instant disease diagnostic (Point-of-care diagnostic), and improvising real-world robotics”)? Also, how much other coverage has this work gotten so far?
(My response)
Neuromorphic devices are designed to do multiple processing at low power – just like an animal brain. Because of their organic nature, they can be integrated with nerve cells. This allows for a nerve cell (or group of nerve cells) to interface with a computer. Since they are able to receive, process, and execute information parallelly, in the future, it can have advantages in diagnosing bodily conditions immediately, instead of sending samples far away for examinations. I haven’t seen its coverage in any mainstream news outlets so far.
