A customized grid environment, based on gym-simplegrid, for reinforcement learning featuring obstacles, path planning, and goal-seeking behavior.
In these examples, the blue circle represents the agent, the red circle is the goal, and the cyan path shows the optimal route calculated by the D* Lite algorithm. Black squares indicate obstacles that the agent must navigate around.