
An enhanced covert navigation framework that leverages LiDAR data, height maps, cover maps, and potential threat maps, along with offline reinforcement learning, to enable autonomous robots to navigate efficiently while minimizing exposure to threats and maximizing cover utilization in complex outdoor environments.


coremsg

enhanced-covert-maneuver-planning-using-offline-reinforcement-learning-for-autonomous-robots-in-complex-outdoor-environments


Enhanced Covert Maneuver Planning using Offline Reinforcement Learning for Autonomous Robots in Complex Outdoor Environments


title_rewrite


TopoNav, a novel framework, integrates active topological mapping, hierarchical reinforcement learning, and intrinsic motivation to enable efficient and autonomous exploration and navigation in unknown environments with sparse rewards.


efficient-topological-navigation-for-autonomous-exploration-in-sparse-reward-environments


Efficient Topological Navigation for Autonomous Exploration in Sparse Reward Environments
