Sense Aeronautics’ Solar Panel Inspections solution applies advanced AI models to automatically detect, classify, and geo-locate defects in photovoltaic installations. Using RGB, thermal, radiometric, and electroluminescence imagery acquired from drones or stationary systems, it transforms raw visual data into actionable insight, enabling operators to identify issues such as cracks, hotspots, and surface degradation before performance loss occurs. The software processes data in the cloud, on local servers, or directly at the edge, supporting flexible deployment across varied inspection environments.
Designed for scalable operations and seamless integration with unmanned workflows, the system ensures consistent analysis across large solar farms. Automated data ingestion and intelligent image interpretation remove the need for manual review, cutting inspection time while improving defect detection reliability. Structured outputs and reporting tools provide clear summaries and integrations for maintenance and asset management platforms.
Specifications:
| Input Sources | RGB, thermal, radiometric, and electroluminescence imagery |
| Input Formats | RAW, JPEG, PNG, TIFF |
| Supported Defects | Cracks, hotspots, PID, delamination, scratches, soiling |
| Reporting | JSON via API, webhooks, HTML, and PDF reports |
| Deployment | Cloud, local, or edge environments |
| Detection Probability | > 90% on major defect types |
| False Alarm Rate | < 10% under recommended conditions |
| Detectable Defect Size | 10 to 40 pixels depending on defect type |
