EnCoMP: Enhanced Covert Maneuver Planning with Adaptive Threat-Aware Visibility Estimation using Offline Reinforcement Learning

--Autonomous robots operating in complex environments face the critical challenge of identifying and utilizing environmental cover for covert navigation to minimize exposure to potential threats. We propose EnCoMP, an enhanced navigation framework that integrates offline reinforcement learning and our novel Adaptive Threat-A ware Visibility Estimation (A T A VE) algorithm to enable robots to navigate covertly and efficiently in diverse outdoor settings. A T A VE is a dynamic probabilistic threat modeling technique that we designed to continuously assess and mitigate potential threats in real-time, enhancing the robot's ability to navigate covertly by adapting to evolving environmental and threat conditions. Moreover, our approach generates high-fidelity multi-map representations, including cover maps, potential threat maps, height maps, and goal maps from LiDAR point clouds, providing a comprehensive understanding of the environment. We train a Conservative Q-Learning (CQL) model on a large-scale dataset collected from real-world environments, learning a robust policy that maximizes cover utilization, minimizes threat exposure, and maintains efficient navigation. We demonstrate our method's capabilities on a physical Jackal robot, showing extensive experiments across diverse terrains. These experiments demonstrate EnCoMP's superior performance compared to state-of-the-art methods, achieving a 95% success rate, 85% cover utilization, and reducing threat exposure to 10.5%, while significantly outperforming baselines in navigation efficiency and robustness. Autonomous navigation in complex environments is a critical capability for robots operating in various applications, such as military reconnaissance [1], search and rescue missions [2], and surveillance operations [3]. These scenarios pose unique challenges for robots, requiring them to accurately perceive the environment, identify potential cover, and adapt their navigation strategies to minimize exposure to threats while efficiently reaching the goal. Jumman Hossain, Abu-Zaher Faridee, and Nirmalya Roy are with the Department of Information Systems, University of Maryland, Baltimore County, USA. Jade Freeman and Timothy Gregory are with the DEVCOM Army Research Lab, USA. Theron Trout is with Stormfish Scientific Corp. Developing robust navigation strategies that can effectively navigate in these environments while maintaining covertness is a challenging task, as it requires accounting for various uncertainties, dynamic obstacles, and environmental factors. Existing approaches to covert navigation often rely on pre-defined environmental models [4], [5] or supervised learning techniques [6], [7] that require extensive manual annotation and labeling of traversable terrain.