To ensure long-term space missions, an autonomous localization and mapping system for lunar rovers is demanded. While the target-oriented localization and mapping problem can be solved through state-of-the-art methods, they greatly rely on human-in-loop remote operations, posing several challenges for the visual system of a rover when operating in a distant, unknown, and feature-sparse lunar environment. This paper presents a SAM-augmented target-oriented SLAM framework that enables rovers to estimate the relative distance to the target on the lunar surface, thus ensuring the safety of the exploration task. Based on the proposed point-prompted object instance extraction pipeline, object correspondences are firstly predicted in the middle-end of Lo-SLAM, where reliable semantic constraints are robustly associated cross image frames. We then maintain camera-object relative positioning between the camera and target in the visual odometry of Lo-SLAM. Meanwhile, a multi-level mapping and representation framework is proposed to keep the target explicit and characterized in different subtasks. Extensive experiments are conducted on our dataset, SePT (i.e., Stereo Planetary Tracks). Results show that the proposed Lo-SLAM is validated on challenging lunar scenarios with dramatic viewpoints and object scale changes. The average pose errors are 0.36 meters in centroid and 0.27 meters in scale, and the average object-centric trajectory error is 0.49% or so. An open-source dataset has been released at https://github.com/miaTian99/SePT_Stereo-Planetary-Tracks.
1 January 2026 in IEEE Transactions on Geoscience and Remote Sensing
Spacecraft autonomous localization is a fundamental capability for space missions, including rover planetary surface exploration, satellite on-orbit servicing and lander precision landing tasks. While conventional approaches predominantly depend on ground-based observatories, their lack of autonomy and scalability limits their applications in deep-space exploration tasks. This paper presents Deep Visual Feature-Driven Map-Assisted Localization (DVD-MapAL), a novel method that combines visual odometry with prior map constraints to establish a self-contained localization system. The proposed architecture introduces two key innovations: (1) a coordinate-constrained pose optimization module that dynamically corrects cumulative errors in visual odometry through geometric consistency verification, and (2) a lightweight deep feature descriptor with a spatial inhibition mechanism enabling robust projection estimation between orbital imagery and pre-existing maps. We release a comprehensive simulation dataset containing 11,721 photorealistic images with six-degree-of-freedom annotations, generated through Unreal Engine and AirSim platforms. Experimental evaluations demonstrate DVD-MapAL’s superior performance, achieving 23.64% lower mean positional error than ORB-SLAM3 in spacecraft scenarios. Further validation on real-world datasets reveals 17.89% improvement over conventional visual localization methods, particularly in feature-deprived extraterrestrial environments. The open-source implementation and benchmark dataset will be made publicly available.