Minimalist SLAM algorithms for unmanned approaches in constrained spaces

Due to the narrow tunnel width, low illumination, and complex turns in underground mines, high requirements are placed on the accuracy and stability of autonomous navigation systems. In this study, a modular navigation and control system was developed for articulated unmanned transport vehicles to realize intelligent driving in such constrained environments. The system uses trajectory markers based on UWB for initial path planning, and integrates real−time LiDAR data for positioning correction. A polar coordinate solver, coupled with a safety−boundary adjustment strategy, was designed to compute steering angle and angular velocity. Control instructions were simplified through an adaptive PID controller. The path planning logic and control output model of a minimalist SLAM algorithm were also systematically established to clarify the core functional mechanisms. Field validation was carried out in the ore haulage roadway of Huaxi Tongkeng Mine. The experimental results showed that the proposed system achieved continuous autonomous navigation under no−light conditions, with an average path deviation of less than 0.35 m, and stable 24−hour unmanned operation. The system achieved 59.65% of the efficiency of manual operation. This study is expected to provide a low−cost and practical solution for intelligent transformation of mining under complex geological and constrained spatial conditions.