For the future exploration of our solar system, missions to Small Solar System Bodies (SSSBs), such as asteroids or comets, are a promising and scientifically important area. The current missions exploring SSSBs require extensive human monitoring and processing to ensure safe operations. By incorporating more onboard autonomy, spacecrafts can explore SSSBs more efficiently and increase scientific output. The Astrone project posed a novel concept of a low-altitude hovering vehicle directly on the surface of an SSSB. This paper presents the AI Mapping, as a part of the extension of the Astrone project's navigation concept with additional AI-based algorithms. The AI Mapping consists of two parts. First, an overview of the AI-based Light imaging, detection, and ranging (LiDAR)/Camera data fusion will be given. With this method, a resolution increase of low-resolution flash-LiDAR data by a factor 8×8 was achieved. Second, the obtained high-resolution data was used in addition to the corresponding wide-angle monocular 2D grayscale images of the irregular and unstructured surface of the SSSB to pre-identify possible landing sites. The Convolutional Neural Network (CNN)-based approach was trained and tested with our artificially generated data set. Different versions of the Deeplabv3+ with certain ResNet and ResNet-RS backbone nets were compared. The results show that hazardous areas such as rocks, boulders, and craters on the surface could be detected successfully. With this method, it was possible to achieve an Intersection over Union (IoU) of 67.45 for the semantic segmentation task.