Substation equipment inspection plays a vital role in the reliability of power supply. The traditional manual inspection method is inefficient and heavy workload, and it is difficult to achieve ideal results. Robot inspection has become the future trend of smart grid development. The real-time localization and map creation of substation inspection robot are studied. A SLAM method of inspection robot based on adaptive Monte Carlo algorithm is proposed. Considering the dynamic environment of substation and the complexity of inspection robot′s characteristics, the two-dimensional map creation and real-time localization of robot are realized. The simulation results show that the SLAM positioning accuracy of inspection robot based on AMC algorithm is higher, and the anti-jamming ability for different process noise is stronger. Therefore, it is more suitable for SLAM problem solving and practical application of substation inspection robot.