Photovoltaic Mapping

Photovoltaic (PV) mapping uses remote sensing imagery and machine learning, primarily deep learning models, to automatically identify and locate rooftop solar panels, addressing the critical need for accurate data on distributed PV generation. Current research focuses on improving the reliability and accuracy of these models, particularly by mitigating the impact of variations in image acquisition conditions and exploring different levels of training data supervision, including weakly and minimally supervised approaches. This work is crucial for grid management and energy system planning, enabling better integration of renewable energy sources and more efficient energy distribution.