To meet the demand for intelligent and unmanned development in thermal power plants, an intelligent inspection system has been designed. This system efficiently performs inspection tasks and monitors the operational parameters of key equipment in real-time. The collected data is uploaded to the monitoring center, allowing operation and maintenance personnel to access equipment information promptly. Data analysis is used to provide fault warning and diagnosis for critical equipment. The system employs the Pure Pursuit algorithm, which effectively avoids obstacles and ensures path continuity and stability. Simulation results show that the Pure Pursuit algorithm significantly improves the navigation accuracy and task efficiency of the inspection robot, ensuring the reliability of thermal power plant inspections.