An autonomous inspection robot for extra-high voltage substations

Extra High Voltage Centers are critical to power transmission, yet their maintenance still relies on manual, time-based inspections with handheld thermal cameras. This approach is labor-intensive, costly, and prone to missing early faults. In this paper we present ENORASI, an autonomous robotic platform for condition-based inspection of EHVCs. ENORASI integrates 3D LiDAR-based mapping, visibility scoring, and traversability analysis to enable safe and efficient navigation in complex substation terrain. A decision support system enables smart scheduling and report generation, while an optimal path planner solves a Generalized Traveling Salesman Problem for efficient routing. Component faults are detected by combining RGB-based object recognition with thermal image analysis. Initial deployments demonstrate that ENORASI can increase inspection frequency, reduce operational costs, and support the transition from time-based to condition-based maintenance. These results validate ENORASI as a scalable and effective alternative to manual inspection for critical power infrastructure.