acotsp: Ant Colony Optimization for the Travelling Salesperson Problem
Note: this package is under heavy developement in the moment.
How it works
The idea behind ant colony algorithms are observations of real ant colonies: ants leave their den, choose a direction randomly and search for food. On their way each ant leaves a pheromone trace, which evaporates slowly. If an ant found something to eat, it heads back to the den. Other ants start searching for food, choosing a random direction influenced by the amount of trace on it. This way short pathes which are used often get an even higher pheromone concentration since more ants use it and longer pathes pheromone concentration evaporates. In that manner sooner or later all the ants will use this ant highway, the socalled ant trail.
What does the package offer?
ACO algorithms transfer this natural collective intelligence into algorithms for solving hard computational problems. This package impelements the Ant Colony Optimization (ACO) framework as an optimizer for the popular Travelling Salesperson Problem (TSP).
Currently there is only this developement version of netgen. The package will be available on CRAN soon. To install the current developement version of the package, install the devtools package by Hadley Wickham, load it and type the following line to a R session:
Coming soon ...
Please address questions and missing features about the acotsp package to the author Jakob Bossek firstname.lastname@example.org. Found some nasty bugs? Please use the issue tracker for this. Pay attention to explain the problem as good as possible. At its best you provide an example, so I can reproduce your problem.