Autonomous Planetary Rover Navigation and Crater‐Based Map Augmentation for Path Planning and Rock Localization Using Space <scp>ROS</scp>

ABSTRACT Navigating planetary surfaces is a critical task for autonomous rovers, where unstructured terrains, craters, and dynamic obstacles present significant challenges. Traditional SLAM‐based navigation systems often struggle with unanticipated obstacles and cannot adapt effectively to changing environments. This paper introduces an innovative approach to planetary rover navigation, utilizing ROS2, the NAV2 stack, and Visual SLAM techniques for real‐time localization and path planning. A key innovation is the integration of crater‐based map augmentation, where craters are detected in real‐time to enhance map accuracy and improve navigation decision‐making. The proposed system is evaluated through simulation, where it demonstrates superior performance in terms of localization accuracy and obstacle avoidance compared to traditional SLAM and teleoperation methods. Experiments show that the system significantly reduces localization drift, improves path planning efficiency, and ensures robust navigation in low‐texture planetary terrains. The results suggest that this approach can be a pivotal advancement for future autonomous planetary exploration missions.