This repository contains the source code for the sferes2 CMOEA module, which is a multiobjective evolutionary algorithm designed to solve multimodal problems. The accompanying publication is:
Huizinga J, Clune J. (2018). "Evolving Multimodal Robot Behavior via Many Stepping Stones with the Combinatorial Multi-Objective Evolutionary Algorithm". arXiv:1807.03392.
If you use this software in an academic article, please consider citing:
@article{huizinga2018evolving,
title={Evolving Multimodal Robot Behavior via Many Stepping Stones with the Combinatorial Multi-Objective Evolutionary Algorithm},
author={Huizinga, Joost and Clune, Jeff},
journal={arXiv preprint arXiv:1807.03392},
year={2018}
}
This CMOEA module is intended to be used with the sferes2 framework. In addition, it depends on two other sferes2 modules, the datatools module and the nsgaext module.
First, download the sferes2 framework (you can skip this step if you already have sferes2 installed):
git clone https://github.com/sferes2/sferes2.git
Next, download the two required modules by navigating to the modules inside sferes2, and cloning the two dependecies.
cd sferes2/modules
git clone [email protected]:JoostHuizinga/nsgaext.git
git clone [email protected]:JoostHuizinga/datatools.git
Laslty, download the CMOEA module.
git clone [email protected]:Evolving-AI-Lab/cmoea.git
You are now ready to build experiments with CMOEA! To run the example experiment provided in this repository, first navigate to the sferes2 root folder.
cd ..
Now copy the example experiment to the experiments folder:
cp -r modules/cmoea/cmoea_example exp
And configure and build the experiment:
./waf configure
./waf --exp cmoea_example
If the build was successful, you can now run the experiment with:
build/exp/cmoea_example/maze_rls_cmoea
The code in this repository is licensed under the MIT License.
If you have questions/suggestions, please feel free to email or create github issues.