Intellectual Property India Publishes Patent Application for A Digital Twin Framework For Federated Learning-Based Predictive Irrigation Scheduling Using Soil Moisture Iot Networks And Satellite Remote Sensing Data Fusion.

MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202641089453 A) filed by Mrs. Shashikala Reddigari; Mr. Ravinder Oranganti; and Dr. K. Ravi Kumar on July 22, 2026, for A Digital Twin Framework For Federated Learning-Based Predictive Irrigation Scheduling Using Soil Moisture Iot Networks And Satellite Remote Sensing Data Fusion..

Inventors include Mrs. Shashikala Reddigari; Mr. Ravinder Oranganti; and Dr. K. Ravi Kumar.

The application for the patent was published on July 31, 2026, under issue no. 31/2026.

Abstract: ABSTRACT OF THE INVENTION: A Digital Twin (DT) framework for predictive irrigation scheduling is disclosed. The framework integrates Internet of Things (IoT) soil moisture sensor networks with satellite remote sensing data (NDVI, LST) to create a comprehensive data acquisition layer. A privacy-preserving federated learning model is employed to synthesize data from multiple farms, training a robust global model without compromising sensitive agricultural data. Each farm operates as a local node, training a model on its own data and sharing only model parameters with a central aggregation server. The resulting global model powers a Digital Twin, a dynamic virtual replica of the farm field. This digital twin simulates future soil moisture conditions and analyzes various irrigation scenarios to generate precise, data-driven recommendations for water application. The invention overcomes the limitations of centralized machine learning models and single-source data systems by offering a scalable, secure, and highly accurate predictive tool. The system is configured to provide actionable irrigation schedules (time, duration, zone) to a user interface, significantly reducing water usage (demonstrated by up to 38.5% savings in a case study) while increasing crop yield. The framework's architecture ensures continuous improvement, as the federated model learns from new data over time without requiring a centralized database of raw farm data. This invention addresses critical challenges in modern agriculture, including water scarcity, resource optimization, and the need for privacy-preserving data analytics. It is adaptable to various crop types, soil conditions, and climatic zones.

Disclaimer: Curated by HT Syndication.