FORCE Physical Reservoir Computing Uses Living Neurons to Teach Robots to "Think" Like Humans Do

Driven by living neurons and with no on-board sensors, this FORCE-powered robot proved capable of solving a maze.

Researchers at the University of Tokyo's Graduate School of Information Science and Technology have come up with a means for teaching robots to think more like people — using a physical reservoir computing approach dubbed FORCE: First-Order Reduced and Controlled Error learning.

""I, myself, was inspired by our experiments to hypothesize that intelligence in a living system emerges from a mechanism extracting a coherent output from a disorganized state, or a chaotic state," co-author Hirokazu Takahashi, associate professor of mechano-informatics, explains. "A brain of [an] elementary school kid is unable to solve mathematical problems in a college admission exam, possibly because the dynamics of the brain or their 'physical reservoir computer' is not rich enough. Task-solving ability is determined by how rich a repertoire of spatiotemporal patterns the network can generate."

This robot can solve a maze through trial-and-error — thanks to a package of living neurons. (📹: Yada et al.)

That's the idea behind physical reservoir computing, or PRC, and the FORCE system developed by Takahashi and colleagues — implemented using a living neuronal culture.

"We conducted embodiment experiments with a vehicle robot to demonstrate that the coherent output could serve as a homeostasis-like property of the embodied system, which could result in the development of problem-solving abilities," the team explains. Our PRC embodiment was characterized as having a linear readout from neural activities and was substantially different from conventional 'Braitenberg vehicle-type' embodiment of a living neuronal culture in which sensory-motor coupling was optimized through the Hebbian learning."

The core of the concept: Using electrical impulses and neuronal errors to steer the robot through a maze — despite the robot's complete lack of sensors, relying instead entirely on impulses created by a trial-and-error approach.

"In summary, we used a living neuronal culture as PRC and implemented FORCE learning to produce a coherent signal output from a spontaneously active reservoir. The output signal served as a homeostasis-like property, enabling the embodied robot to solve a maze. Our results suggest that the homeostatic property generated from the internal feedback loop in the system plays an important role in employing computational resources for task solving in biological systems."

The work, which the team hopes may lead to breakthroughs in understanding how the brain works and the development of neuromorphic computing systems for robotics and beyond, has been published under open-access terms in the journal Applied Physics Letters.

Gareth Halfacree
Freelance journalist, technical author, hacker, tinkerer, erstwhile sysadmin. For hire: freelance@halfacree.co.uk.
Latest articles
Sponsored articles
Related articles
Latest articles
Read more
Related articles