Simultaneous Localisation and Mapping (SLAM) is an important technology that enables autonomous robots to navigate independently and create detailed, real-time maps of their surroundings. This capability is crucial in underground mining settings, where it enhances operational efficiency and significantly strengthens safety in these subterranean environments. SLAM algorithms typically rely on identifying and using distinct geometric features, such as plane and edge features, to understand the spatial layout of the environment and to track how it evolves as the sensor moves through it. In underground roadways, the scarcity of noticeable edge features, coupled with the repetitive nature of walls and corridors, poses a significant challenge, making it difficult for these systems to distinguish one area from another. Furthermore, the absence of global navigation satellite system (GNSS) signals in underground adds more challenges, leading to the accumulation of errors in both the mapping and localisation.