Home » News » Artificial GPS brain cells helps robot to find its way

Artificial GPS brain cells helps robot to find its way

Researchers in Singapore have given a robot a simulated version of the brain cells used by animals to build a mental map of their environment.

The interplay of two specific types of neurons in the brain are responsible for our ability to navigate our surroundings. One type of navigation cells, place cells, was identified by John O’Keefe in the 70s, when he observed that certain cells in the brain fired whenever a mouse passed a specific location. Further discoveries were made in 2005, when May-Britt and Edvard Moser found that other cells, grid cells, were activated when an animal arrived at determined locations on a triangular grid of reference. These two types of cells work in conjunction with other types of cells to process sensory information and give us an innate sense of the world around us. The discovery of these cells earned Keefe, May-Britt and Moser the Nobel Prize for Medicine in 2014.

Scientists have found a way to give a robot similar arrangement of cell enabling it to find its way around

Now scientists have found a way to give a robot a similar arrangement of cells to enable it to find its way around. By simulating the two types of ‘navigation’ cells, the researchers passed on these location and navigation skills to a small-wheeled robot. The team at the Agency for Science, Technology and Research used software to create a simple two-dimensional model of the cells, rather than simulate the cells physically, as they believed that these artificial cells had the ability to provide an adaptive and robust mapping and navigation system.

This research is important as it shows that it’s possible to instruct machines to mimic some of the more complex brain activity. While artificial neural networks are already used to train robots to perform simple tasks such as object recognition, so far these networks have not been able to accurately reflect the complexities of a human or animal brain.

The team in Singapore allowed the robot to roam around a 35 square metre office, and were able to verify that its artificial set of place and grid cells worked in the same way as biological grid and place cells would. While the artificial navigation system doesn’t work as well as the conventional system yet, further research is need to more fully understand how the biological cells function before they attempt to improve it.

What’s Next?

Looking for a job within the Neurological conditions care industry? See our Careers page for exclusive job opportunities at some of the best care facilities.

About dani