5
Researchers at the University of Technology in Malaysia have created an image dataset to help develop autonomous robots that clean solar panels. The dataset contains 6,200 annotated images showing clean panels, simulated dirt at five contamination levels, and physically dirty panels. Researchers can use these images to train and evaluate robotic vision systems that recognize panel conditions, potentially helping robots clean solar panels more effectively.
