Researchers at the National Institute of Technology (NIT) Rourkela have developed an Artificial Intelligence-powered system to clean solar panels more efficiently, targeting only panels that need cleaning to save water and labour. The technology was developed by Prof. Arun Kumar, Assistant Professor, Prof. Bibhudatta Sahoo, Professor, and research graduates Dr Lopamudra Hota and Dr Biraja Prasad Nayak from the Department of Computer Science and Engineering. The team has secured an Indian patent titled “Federated Learning-based Autonomous System and Method for Monitoring and Cleaning Solar Plant”. The system combines AI, federated learning and autonomous cleaning. According to the researchers, solar farms in dry and dusty areas can lose up to 40% of their energy generation because of dust, bird droppings, industrial pollutants and other material settling on the panels. Many existing systems rely on labour and water and often clean panels on a fixed schedule, wasting resources on panels that are already working well. The new system continuously checks panel condition, detects faults, studies real-time conditions and recommends which panels need cleaning. It uses Federated Learning, which allows devices to learn from data without sending all original information to a central server. The technology has so far been tested through simulations and is currently at Technology Readiness Level 3, or TRL-3.