MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641105411 A) filed by Senthilkumar; Mrs. Prabha Rani P; Hemanth Raj Gudisa; R. Dhamodharan; M. Banu Priya; A. Ganesan; Dr. B. R. Senthil Kumar; and Dr V. Jayalakshmi on September 01, 2026, for Deep Learning-Based Autonomous Water Management And Vegetation Control System.
Inventors include Senthilkumar; Mrs. Prabha Rani P; Hemanth Raj Gudisa; R. Dhamodharan; M. Banu Priya; A. Ganesan; Dr. B. R. Senthil Kumar; and Dr V. Jayalakshmi.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: A deep learning-based autonomous system for simultaneous water management and vegetation control in agricultural fields is disclosed, comprising a microcontroller- based multi-parameter environmental sensing arrangement and a camera for capturing field images. An edge artificial-intelligence device executes an object-detection model on the images to identify crop and weed regions and determine weed location and density, while a sequential deep-learning model predicts water requirement from environmental data. A fusion module combines the sensor-derived and visual features into a fused field-state representation from which a joint decision engine, applying a multi-objective cost function, determines a combined irrigation and vegetation-control decision, executed by an actuation arrangement comprising a solenoid valve and pump for drip irrigation and a precision actuator directed to detected weed locations. A closed-loop feedback arrangement verifies outcomes and fine-tunes the models. The system reduces water and input consumption, improves crop-health outcomes over threshold-based systems, and is deployable on low-cost edge hardware for fields with limited connectivity.
Disclaimer: Curated by HT Syndication.
