<p>This review examines decision-making and planning technologies for unmanned driving in mining environments. It starts by analyzing how mining conditions (e.g., low light and high dust) affect perception, decision-making, and planning. By reviewing unmanned driving technologies across different scenarios (urban roads and open-pit mines), this study evaluates the applicability and limitations of multi-sensor fusion, modular decision-making frameworks, and dynamic path planning in mining. Key bottlenecks identified include sensor robustness, real-time performance, and path optimization adaptability. For example, sensor performance degradation impacts perception accuracy, and the real-time requirements versus computational resources trade-off restricts system enhancement. To address these, this review suggests optimizing sensor fusion algorithms, boosting decision-making intelligence, and improving real-time responsiveness. Future research directions highlighted are enhancing multi-sensor fusion accuracy and robustness, developing dynamic environment–adapted intelligent decision-making algorithms, optimizing communication coverage and reliability, and advancing standardization and scalable applications. These findings provide clear guidance for the development of unmanned driving technology in mining environments.</p>