Graph augmented Deep Reinforcement Learning in the GameRLand3D environment
This addresses the problem of poor generalization in deep reinforcement learning for navigation in vast, procedurally generated 3D environments, though it is incremental as it builds on existing hybrid approaches.
The paper tackles planning and navigation in complex 3D video games with disconnected regions, introducing a hybrid method that combines reinforcement learning with a graph-based planner, resulting in a 20% absolute increase in success rate over an end-to-end agent in unseen large-scale maps.
We address planning and navigation in challenging 3D video games featuring maps with disconnected regions reachable by agents using special actions. In this setting, classical symbolic planners are not applicable or difficult to adapt. We introduce a hybrid technique combining a low level policy trained with reinforcement learning and a graph based high level classical planner. In addition to providing human-interpretable paths, the approach improves the generalization performance of an end-to-end approach in unseen maps, where it achieves a 20% absolute increase in success rate over a recurrent end-to-end agent on a point to point navigation task in yet unseen large-scale maps of size 1km x 1km. In an in-depth experimental study, we quantify the limitations of end-to-end Deep RL approaches in vast environments and we also introduce "GameRLand3D", a new benchmark and soon to be released environment can generate complex procedural 3D maps for navigation tasks.