FGA – Fast Genetic Algorithm

About

This is the home page for the FGA project.

FGA is a simple yet powerful implementation of genetic algorithms which provides many choices for crossover and selection procedures.

It is suitable to solve mathematical problems such as combinatorial optimization ones, as well as to build artificial life simulations.

The library is written in C++, and you can easily incorporate it in your own application, provided that you respect the terms of the license.

A parallel version of the algorithm (using POSIX threads) is included in order to take advantage of multi-processor environments.

The library is in the form of a single C++ header and compiles on both Linux/Unix and Windows platforms.

Download

You can download the latest version from SourceForge.net mirrors: fga-1.4.

The tarball includes the library and an example of its use to solve the "travelling salesman problem".

Thanks to FreeBSD developers, users of this operating system can install FGA directly from ports: FreeBSD port.

Documentation

Documentation for the project can be found at this page: FGA documentation.

Additional resources

A fancier graphical version of the TSP example can be downloaded separately: graphical TSP solver.

The experimental source code page shows some ideas for future development of FGA, as well as code contributed by other people.

A research in the field of robotics at the Cornell Computational Synthesis Lab used FGA among other software (poster).

Contact

Please help us testing and feel free to submit your opinions/bug reports to alessandro.presta@gmail.com.

The author's home page can be found at http://alessandropresta.altervista.org.

Powered by