A robot computer algorithm that copies the navigation functionality of humans and animals

October 7, 2015, Agency for Science, Technology and Research (A*STAR), Singapore
A robot computer algorithm that copies the navigation functionality of humans and animals
This robot uses neural schemes similar to humans to navigate an office environment. Credit: A*STAR Institute for Infocomm Research

A robot with a navigation system that mirrors the neural scheme used by humans and animals to find their way around has been developed by A*STAR researchers.

The human navigation function is operated by two types of brain cells—place cells and . Place cells become active in the brain when we recognize familiar places, while grid cells provide us with an absolute reference system, so we can determine exactly where we are on a map.

The way sailors used to navigate through tracking of relative movement, however, is essential for finding a way through unfamiliar areas, explains Miaolong Yuan from the A*STAR Institute for Infocomm Research team. "A sailor will use cues such as the stars or landmarks to determine where their ship is on a map, and then, as the ship moves, will update its location on the map by observing only speed and direction."

The human brain uses grid cells, which provide a virtual reference frame for spatial awareness to handle this type of relative navigation. Each time we move through and pass one of the virtual grid points that the brain has set up, the respective grid cell becomes active, and we know our relative movement in relation to those coordinates. By using both place and grid cells for navigation, humans and animals are able to accurately move through the environment.

Yuan and the team have implemented the same neural scheme for robots, using computer programs that simulate the activity of place and grid cells in the brain. Crucial to the computational algorithm is the strength of the feedback mechanism between the grid cells and place cells, and the calibration of the visual signals is integral to the map building process of the computer algorithm.

The algorithm was tested in a (see image) that explored a 35 meter x 35 meter indoor office environment. The robot was able to detect loops in the path through the office space and, by using visual cues to recognize areas visited repeatedly, built its own neurological map of the office. The computer navigation system assists the robot in situations where it is lost in a new environment, says Yuan. "Cognitive maps can help the robot when it is lost, because they can provide global topological information of the navigating environment to help the robot localize itself."

Explore further: What rats in a maze can teach us about our sense of direction

More information: "An entorhinal-hippocampal model for simultaneous cognitive map building." Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, 586–592 (2015). www.aaai.org/ocs/index.php/AAA … paper/view/9420/9298

Related Stories

Study highlights brain cells' role in navigating environment

February 5, 2015

A new Dartmouth College study sheds light on the brain cells that function in establishing one's location and direction. The findings contribute to our understanding of the neural mechanisms underlying our abilities to successfully ...

Connecting places causes mental maps to merge

April 23, 2015

Realising how places connect geographically causes local maps in the brain to join, forming one big map which helps with planning future journeys, finds a new UCL study.

Better understanding of mapmaking in the brain

August 9, 2010

"Grid cells," which help the brain map locations, have been found for the first time outside of the hippocampus in the rat brain, according to new research from the Norwegian University of Science and Technology (NTNU). The ...

Brain's GPS system influenced by shape of environment

February 11, 2015

Patterns created by the brain's grid cells, which are believed to guide navigation, are modified by the shape of the environment, according to UCL researchers. This means grid patterns aren't a universal metric for the brain's ...

Recommended for you

What can snakes teach us about engineering friction?

May 21, 2018

If you want to know how to make a sneaker with better traction, just ask a snake. That's the theory driving the research of Hisham Abdel-Aal, Ph.D., an associate teaching professor from Drexel University's College of Engineering ...

Flexible, highly efficient multimodal energy harvesting

May 21, 2018

A 10-fold increase in the ability to harvest mechanical and thermal energy over standard piezoelectric composites may be possible using a piezoelectric ceramic foam supported by a flexible polymer support, according to Penn ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.