Showing posts with label Ant Colony Optimization. Show all posts
Showing posts with label Ant Colony Optimization. Show all posts

Monday, April 1, 2013

Ant Colony Optimization

By (O)CT(O)PUS



A tiny cube, two watch motors to power the wheels, a pair of antennae for light sensors, limited memory and less processing power, scientists have built and tested robotic ants that behave just like a real ant colony.

Individually, each cube is a dumb cube - moving around at random and getting lost. Yet, their secret is in their ability to take cues from other cubes – just like a swarm of insects.

Just like real ants, robotic cubes leave a trail that others can follow – not a pheromone trail – but a trail of light that will stimulate other cubes to follow the same path.

As Simon Garnier of the SwarmLab explains: "[The robots] have two antennae on top, which are light sensors. If more light falls on their left sensor, they turn left; and if more light falls on the right sensor, they turn right."

"If there are two possible paths from A to B and one is twice as long … the ants [or] robots start using each path equally.

"Because ants taking the shorter path travel faster, the amount of pheromone (or light) deposited on that path grows faster, so more ants use that path."  The result is a positive feedback loop that causes them collectively to take the shortest path 2pointB.

Limited memory, less processing power, blathering to a base of robotic cubes, ant colony optimization sorta reminds me of Republicans.