the easy to use artificial life simulation environment
 
 
Go to file
Enrico Fasoli dd76b3acc3 moved from single precision floating point to double precision 2016-03-01 15:50:48 +01:00
.nb-gradle big update with carnivores 2015-07-02 21:42:11 +02:00
core moved from single precision floating point to double precision 2016-03-01 15:50:48 +01:00
desktop moved from single precision floating point to double precision 2016-03-01 15:50:48 +01:00
gradle/wrapper first commit 2015-06-28 01:27:17 +02:00
.gitignore big update with carnivores 2015-07-02 21:42:11 +02:00
.nb-gradle-properties first commit 2015-06-28 01:27:17 +02:00
README.md Update README.md 2015-08-10 15:20:01 +02:00
build.gradle first commit 2015-06-28 01:27:17 +02:00
gradle.properties first commit 2015-06-28 01:27:17 +02:00
gradlew first commit 2015-06-28 01:27:17 +02:00
gradlew.bat first commit 2015-06-28 01:27:17 +02:00
settings.gradle first commit 2015-06-28 01:27:17 +02:00

README.md

AIrium

AIrium Short GIF

This project aims to build an interactive simple artificial life simulation that uses genetic algorithm and fitness evaluation to improve, over different generations, the performance of the agents.

For now, agents are evaluated based on how much they eat. Using default settings you can see them evolve to be herbivores, but by tweaking it is possible to get carnivores or scavengers.

In the GIF you can see an environment similar to the default, but smaller. Most creatures are pretty efficient herbivores with a few showing slight carnivore behavior. This was obtained in only 5 generations of training, in about two minutes on an average CPU. You can see a GIF with a totally different outcome here, it shows some creatures which evolved to eat food by stopping when they encounter it and always travelling in a straight line. This was obtained in a little more time with a slightly different configuration, showing that a lot can change just by running different simulation even with the same settings.

Since version 0.5 it is possible to save your best creatures to disk so you can reintroduce them to different environments later and avoid losing very efficient agents.

Features

  • Artifical Life creatures with simple brains
  • Artificial selection to ensure the evolution of the creatures over time
  • Many configurable settings to create different environments (still a little limited, but I'm working on it)
  • Easy to use with working default settings for beginners, highy configurable and fully open source for hackers!
  • Save and Load your configuration and your creatures
  • Every creature has a unique, persistent name generated using its brain topology, so on every computer the same creature will have the same name
  • Multithread design, resizeable window, configurable simulation speed, multiplatform support, free and open source!

Screenshots and videos

You can see some footage from version 0.4 in this imgur album

Download

Visis the releases page to download the latest version.

It works on Linux, OSX and Windows. You will need Java 8:

  • On Linux you can find the OpenJDK JRE version 1.8 on your package manager repositories
  • On Windows and OSX you can get the Oracle Java VM from here

Planned features

  • More behavioral variety
  • More sensors for the creatures
  • Ability to save and load simulation states
  • More in the issues page

Controls

  • The Configuration page in the GUI will let you tweak many parameters and pause the simulation
  • You can drag with the mouse to move the camera and zoom with the mouse wheel

Development

You will need a Java Development Kit (JDK) version 1.8 (Java 8) or greater. Both the Oracle and OpenJDK distributions have been tested.

To build from source:

  • on Linux and OSX, make sure you have a Java Development Kit (JDK) version 1.8 or greater, change directory AIrium and then use gradle to build the software and generate an executable, like this: ./gradlew :desktop:dist. The generated jar executable will be at AIrium/desktop/build/libs/desktop-1.0.jar. You can find information about the gradle command line interface here.
  • on Windows instead of ./gradlew :desktop:dist you'll need to use gradlew.bat :desktop:dist

To develop and hack on the code with an IDE, I suggest you install Netbeans 8 and follow Libgdx's guide to set it up for Libgdx development (mobile platform are ignored for now so you won't need the Android SDK). You can find further explanation in Libgdx's wiki.

License

Copyright (c) 2015 Enrico Fasoli (fazo96)

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.