An Organic "Human-Like Brain" Gives This LEGO Robot Maze-Solving Smarts

Inspired by nature, this organic robot "brain" links the sensors to the motors as closely as possible.

Researchers at the Eindhoven University of Technology (TU/e) and the Max Planck Institute for Polymer Research have given a robot an organic "human-like brain" for maze-solving puzzles β€” and believe it heralds great things for future neuromorphic devices.

"Just as a synapse in a mouse brain is strengthened each time it takes the correct turn in a psychologist's maze, our device is 'tuned' by applying a certain amount of electricity," Imke Krauhausen, principal author of the team's paper, explains. "By tuning the resistance in the device, you change the voltage that control the motors. They in turn determine whether the robot turns right or left."

This simple robot uses an organic "human-like brain" to solve a hex-based maze. (πŸ“Ή: TU/e)

That device, housed in a LEGO Mindstorms EV3 wheeled robot, is an organic "brain," which is designed to operate much like a human brain β€” though significantly simpler β€” combined with novel integration between the robot's sensors and its motors.

"This sensorimotor integration, in which sense and movement reinforce one another, is also very much how nature operates," Krauhausen explains, "so this is what we tried to emulate in our robot."

Prior to the maze attempt, the robot's "brain" is pre-trained using visual stimuli that act as navigation cues; as it progresses through the maze, the neuromorphic circuit β€” which has created an association between the sensors and motors β€” is able to navigate autonomously, taking 16 steps to figure out a successful route to the exit.

A neuromorphic circuit links sensors to motors in a way inspired by nature. (πŸ“·: Krauhausen et al)

The "thinking" robot proved adept at solving a hex-based maze, but its creators are cautious about the complexities involved in extending its capabilities. "Our robots still rely partly on traditional software to move around," Krauhausen admits of the work still to be done.

"And for the neuromorphic robots to execute really complex tasks, we need to build neuromorphic networks in which many devices work together in a grid. That’s something that I will be working on in the next phase of my PhD research."

The team's work has been published under open-access terms in the journal Science Advances.

ghalfacree

Freelance journalist, technical author, hacker, tinkerer, erstwhile sysadmin. For hire: freelance@halfacree.co.uk.

Latest Articles