researchers converted a mathematical classicPuzzle in a maze of ants.
Image: The University of Sydney
A colony of ants is probably the last place they can expect to find a math whiz, but researchers at the University of Sydney has shown that humble Ant is not only capable of difficult mathematical problems, but it is even able to do what few computer algorithms can - adapt the optimal solution to adapt to a problem of evolution.
These findings, published in the journal of Experimental Biology, deepen our understanding of the animals how simple can overcome a complex and dynamic problems in the nature and the desire to help it professionals to develop even better software to solve logistical problems and maximize efficiency in many human industries.
Using an original technique, Chris Reid and Madeleine Beekman School of biological sciences, associate professor in collaboration with Professor David Sumpter of Uppsala University in Sweden, verified whether Argentine ants (Linepithema humile) can resolve dynamic optimization problem by converting the Towers of Hanoi math classic puzzle in a maze.
Find the path most effective thanks to a busy network is a common challenge encountered by delivery drivers, telephone routers and engineers. To solve these problems of optimization software, it professionals have often sought inspiration from the colonies of ants in nature - creation algorithms that simulate the behavior of ants are more efficient roads in their nests in food sources following mutually volatile pheromone trails. The most commonly used of these algorithms inspired ant is known as Ant Colony Optimization (CoA).
"Although inspired by nature, these computer algorithms often do not represent the real world because they are static and designed to solve a single problem and immutable, says lead author Chris Reid, behavioural and social insect Genetics laboratory doctoral student."
"But the nature is full of unpredictability and a solution does not match all." If we turned the Ant to see how well their problem-solving skills to respond to change. They are attached to a single solution, or they can adapt? »
Researchers tested Ant using the three-rod version, three-disc of the Towers of Hanoi - a puzzle toy requires players problem moving disks between shoots while complying with certain rules and using less of moves possible. But given that ants cannot move the disks, the researchers to convert the puzzle in a maze where the shortest path is the solution less moves in the puzzle toy. Ants at point of entry of the labyrinth could choose between 32,768 paths possible to go to the source of food, on the other hand, with only two of the shortest paths path, and therefore the optimal solution.
Ants have received an hour to solve the maze by creating a path of intense traffic between their nest and the source of food, then researchers blocked off the coast of paths and opens up new areas of maze to test the dynamic problem solving ability of ants.
After an hour, ants solved the towers of Hanoi by finding the shortest path to the edge of the labyrinth. But when this path was blocked off the coast, ants answered by their original path around the obstacle and the establishment of a long, sub-optimal route curve. But after an additional hour, ants, did the maze by abandoning their sub-optimal route and establishing a road crossing the Centre of the labyrinth on the new optimal route.
But problem solving skills not all colonies were equal: less errors and were faster to solve the maze to ants were naive ants were allowed to explore the labyrinth without food for an hour before testing. This result suggests that "exploratory pheromone" worn down by ants research a new territory is the key to help them adapt to changing conditions.
"Simple Ant recruitment mass have much more complex and more labile problem problems that we thought than ever." Unlike previous belief, Ant pheromone system does not get stuck in a particular path and can not adapt. Having at least two separate pheromones provides much more flexibility and help to find the best solutions in a changing environment. Discover how ants are capable of dynamic problems can provide the new source of inspiration for optimization, algorithms which in turn can lead to better resolve software problems and therefore more effective for human industries. »
Editor's note:Original press release here.
No comments:
Post a Comment