MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641066556 A) filed by Rmk Engineering College on May 27, 2026, for An Ai-Driven Crop Yield Forecasting System With Climate-Adaptive Reinforcement Learning Optimization Framework.
Inventors include Dr. P. Shobha Rani; Dr. S Srijayanthi; Dr. Gladiss Merlin N. R; Dr Shanthi M; and Dr Sandra Johnson.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: The present invention relates to an AI-driven crop yield forecasting system (100) integrated with a climate-adaptive reinforcement learning optimization framework designed to enhance agricultural productivity and decision-making. The proposed system collects agricultural and environmental data through a data collection layer (110) comprising soil sensors, weather monitoring stations, and satellite-based remote sensing sources. These sensors capture important parameters including soil moisture, pH level, nutrient concentration, temperature, rainfall, and humidity. The acquired data are transmitted to a data processing module (120) that performs preprocessing operations such as data cleaning, normalization, and feature extraction to transform raw agricultural data into structured datasets suitable for predictive analysis. The processed information is then analyzed by an AI prediction layer (130) that employs machine learning and deep learning techniques to estimate crop yield under varying climatic and environmental conditions. Machine learning algorithms such as Random Forest and Gradient Boosting are utilized for predictive modeling, while deep learning models such as Long Short-Term Memory (LSTM) networks analyze temporal climate patterns and historical crop datasets to improve prediction accuracy. The predicted yield data are further processed through a reinforcement learning optimization module (140) which includes a reinforcement learning agent capable of generating adaptive agricultural management strategies. The reinforcement learning agent evaluates environmental states such as soil conditions, weather forecasts, and crop growth stages to determine optimal actions including irrigation scheduling, fertilizer application, crop variety selection, and pest management practices. The decision policies generated by the system aim to maximize crop productivity while optimizing resource utilization and minimizing environmental risks. The results are provided through an output layer (150) which generates crop yield forecasts, climate risk alerts, and farming recommendations for farmers, agricultural researchers, and policy makers. The proposed system enables climate-resilient agriculture by combining artificial intelligence, reinforcement learning, and real-time agricultural data analytics, thereby improving precision farming practices and supporting sustainable food production. Keywords AI-driven agriculture, crop yield forecasting, reinforcement learning optimization, precision agriculture, machine learning, climate-adaptive farming, IoT-based agricultural monitoring, deep learning in agriculture, smart farming systems, agricultural decision support system.
Disclaimer: Curated by HT Syndication.
