I am implementing two reinforcement learning algorithms, Q-Learning and SARSA, to develop an intelligent agent navigating through Grid-World and find the optimal path. The Grid-World is a world
consisting of four types of nodes: “Flat”, “Mountain”, “Goal” and “Pitfall”.The agent aims to navigate on the Grid-World and reach a “Goal” node while maximizing the total reward/score. I have grids with multiple goals with same reward values. Is there a simple approach I can take to solve multi-goal qlearning?

A resource suggestion or solution suggestion can be nice.