A personal project for creating a multi-agent simulation that is widely configurable for RL training, etc. Agents can use ray casting to detect obstacle edges and surrounding other agents. A gui for display and control was built with pygame. The gui has the ability to select an agent and see the simulation from the agents sensors.
This project is currently WIP