Watch Video: This robot can find its way home on its own without GPS or mapping

Scientists claim to have developed first walking robot that can explore its environment and find its way home without GPS or mapping, opening new avenues for navigation of automous vehicles.

Researchers from French National Center for Scientific Research (CNRS) took inspiration from desert ants, which are extraordinary solitary navigators, to design AntBot.

Ants use polarised light and ultraviolet radiation to locate mselves in . Cataglyphis desert ants in particular can cover several hundreds of metres in direct sunlight in desert to find food, n return in a straight line to nest, without getting lost.

READ | Watch how this robot learns to recognise human actions and emotions

AntBot copies desert ants' exceptional navigation capacities, allowing it to explore without using Global Positioning System (GPS), researchers said.

It is equipped with an optical compass used to determine its heading by means of polarised light, and by an optical movement sensor directed to Sun to measure distance covered.

Armed with this information, AntBot has been shown to be able, like desert ants, to explore its environment and to return on its own to its base, with precision of up to one centimetre after having covered a total distance of 14 metres.

READ | This robot learns how to play Jenga, but worried humans call it "CHEATING"

Weighing only 2.3 kg, this robot has six feet for increased mobility, allowing it to move in complex environments, precisely where deploying wheeled robots and drones can be complicated.

optical compass developed by scientists is sensitive to sky's polarised ultraviolet radiation. Using this "celestial compass," AntBot measures its heading with precision by clear or cloudy wear. 

AntBot brings new understanding on how desert ants navigate, by testing several models that biologists have imagined to mimic this animal, researchers said.

READ | Amazon is actively building an army of self-walking robots to deliver packs

Before exploring potential applications in aerial robotics or in automobile industry, for example, progress must be made, for instance in how to operate this robot at night or over longer distances.