Bot playing to microRTS (github.com/santiontanon/microrts) and exploiting GHOST (github.com/richoux/GHOST) to solver optimization problems under uncertainty
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data
include/ghost
lib
maps
problem_model
resources
results
src
.gitignore
LICENSE
README.md
help.png
launch.sh
launch_n.sh
timeouts_competition.txt

README.md

Compiling our bot

First, clone the develop brach of our GHOST framework with: git clone --single-branch -b develop https://github.com:richoux/GHOST.git

Enter into the GHOST folder and compile it with the following commands (you need cmake and g++ or clang): ./build.sh (on Linux, do not forget to run "sudo ldconfig" when the build.sh script finished)

Then, build the solver required by our bot:

$> cd problem_model $> make all_rts

It will compile and place the executable "solver_cpp" in the src/ai/poadaptive folder where our bot is.

You can run the script "launch.sh" to be sure everything is (locally) ok!

microrts

microRTS is a small implementation of an RTS game, designed to perform AI research. The advantage of using microRTS with respect to using a full-fledged game like Wargus or Starcraft (using BWAPI) is that microRTS is much simpler, and can be used to quickly test theoretical ideas, before moving on to full-fledged RTS games.

microRTS is deterministic and real-time (i.e. players can issue actions simultaneously, and actions are durative). It is possible to experiment both with fully-observable and partially-observable games. Thus, it is not adequate for evaluating AI techniques designed to deal with non-determinism (although future versions of microRTS might include non-determinism activated via certain flags). As part of the implementation, I include a collection of hard-coded, and game-tree search techniques (such as variants of minimax, Monte Carlo search, and Monte Carlo Tree Search).

microRTS was developed by Santiago Ontañón.

For a video of how microRTS looks like when a human plays see a youtube video

An AI competition was organized aroung microRTS in the IEEE-CIG 2017 conference, and again in 2018. For more information on the competition see the competition website

To cite microRTS, please cite this paper:

Santiago Ontañón (2013) The Combinatorial Multi-Armed Bandit Problem and its Application to Real-Time Strategy Games, In AIIDE 2013. pp. 58 - 64.

Contributions:

The LSI AI was contributed by Alexander Shleyfman, Antonin Komenda and Carmel Domshlak (the theory behind the AI is described in this paper.

Instructions:

instructions image