This project demonstrates the use of SAT solvers to schedule path finding in multi agent systems. The implementation is based on Efficient SAT Approach to Multi-Agent Path Finding under the Sum of Costs Objective
The demonstrator provides a playground for simulating different cell based agent paths.
- UI for adding and managing agents
- Store agent setups in presets to load them later
- Selection of maps including DAO-maps mentioned in the paper
git submodule update --init
Building glucose
cd glucose
mkdir build
cd build
cmake .. -DCMAKE_POSITION_INDEPENDENT_CODE=ON
cmake --build .
Building MAPF plugin (gdext)
cd ..
godot --dump-extension-api
cd godot-cpp
scons platform=linux custom_api_file=../extension_api.json
cd ..
scons platform=linux