Skip to content

Latest commit

 

History

History
90 lines (70 loc) · 3.98 KB

README.md

File metadata and controls

90 lines (70 loc) · 3.98 KB
Raw

README

ARGoS3-AutoMoDe

All useful information about the ARGoS3-AutoMoDe package, including installation and utilization instructions, are regrouped in the following technical report (techrep). Please cite this report if you use the ARGoS3-AutoMoDe package.

Package content

  • bin This empty folder will contain the executable automode_main after compilation.
  • experiments The folder where all experiments configuration files (.argos or .xml) are located
    • chocolate The experiment configuration files of the experiments described in chocolate
    • example The experiment configuration file used as example in the technical report of the package.
  • optimization The folder where one should place files regarding the optimization algorithm used (including grammar generator for the irace algorithm)
    • example The necessary elements to launch irace on a small example described in the technical report.
  • src The source files of ARGoS3-AutoMoDe package.
    • cmake The .cmake files for ARGoS3.
    • core The core classes of ARGoS3-AutoMoDe.
    • modules The behaviors and conditions modules described in chocolate
  • AutoMoDeMain.cpp :: The main procedure to launch ARGoS3-AutoMoDe.

Installation

Dependencies:

Compiling AutoMoDe:

$ git clone https://github.com/demiurge-project/ARGoS3-AutoMoDe.git
$ cd argos3-AutoMoDe
$ mkdir build
$ cmake ..
$ make

Once compiled, the bin/ folder should contain the automode_main executable.

How to use

Run a single experiment:

To run a single experiment, you need to use the automode_main as follows:

usage: automode_main [-r | --readable] [-c | --config-file configfile] [-m | --method method] [-s | --seed seed] [--fsm-config fsmconfig]
    options:
        [-c | --config-file configfile] Path to the .argos|.xml experiment file to use [REQUIRED]
        [-r | --readable] Prints an URL containing a DOT representation of the finite state machine [OPTIONAL]
        [-s | --seed seed] The random seed for the ARGoS3 simulator [OPTIONAL]
        [--fsm-config fsmconfig] the description of the finite state machine in the AutoMoDe format [REQUIRED]

Run the design process

To run the design process you will need the irace package and a fonctional R distribution.

If you are using an MPI capable computing cluster you can use the parallel-irace-mpi located in the optimization/ folder. Otherwise, you can directly launch irace with the command:

irace --exec-dir=EXECDIR --seed SEED --scenario scenario
    options:
        [--exec-dir=execdir] the execution folder
        [--seed seed] the random seed for the optimization algorithm
        [--scenario scenario] the scenario file of the experiment

For example (in the optimization/ folder):

irace --exec-dir=aggregation --seed 1234 --scenario scenario.txt

This will run the optimization algorithm for 200k simulation in order to train for the aggregation mission (see chocolate for details on missions).

References

Bibliography

  • [techrep] Ligot, A., Hasselmann, K., Delhaisse, B., Garattoni, L., Francesca, G., & Birattari, M. (2017). AutoMoDe, NEAT, and EvoStick: implementations for the E-puck robot in ARGoS3. Technical report TR/IRIDIA/2017-002, IRIDIA, Université libre de Bruxelles, Belgium.
  • [chocolate] Francesca, G., Brambilla, M., Brutschy, A., Garattoni, L., Miletitch, R., Podevijn, G., ... & Mascia, F. (2015). AutoMoDe-Chocolate: automatic design of control software for robot swarms. Swarm Intelligence, 9(2-3), 125-152.