The autonomous vehicle that thinks like an ant
September 23, 2008 We've seen numerous examples of science mimicking nature in the developing fields of robotic automation and artificial intelligence in recent times, from robotic fish to leaping micro-bots and mechanical rodents. Now a team of engineers from the University of La Laguna (ULL) in the Canary Islands have applied this thinking to self-steering vehicles using a system based on the way ants navigate between home-base and their food source.
Known as Verdino, the prototype machine is a modified golf cart that uses an onboard camera to sense the road surface and applies an algorithm based on the way ants move to determine its path and steer the vehicle accordingly in real-time.
Ants find their way by following a trail of pheromones left behind by other members of the colony. This technique known as "Ant Colony Optimisation" forms the basis of the algorithms used to control Verdino's path without any driver intervention.
The researchers report promising results from a current trial in which the vehicle is being used as an internal transport system to link 25 housing units and a visitor center in the south of Tenerife. Potential applications for the method include driver-less transport in tourist complexes, shopping centers, industrial estates, or even in vehicles modified for use by elderly or disabled people.