MapExRL: Human-Inspired Indoor Exploration with Predicted Environment Context and Reinforcement Learning

We propose MapExRL, a human-inspired method combining reinforcement learning and map predictions for indoor robot exploration.