MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641096643 A) filed by Vellore Institute Of Technology on August 10, 2026, for A Deep Convolutional System And Method Aided Smart Agricultural Soil Enhancement Rover.
Inventors include Ramya. G; and Chaitanya Akshaya Bakshi.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: ABSTRACT The present invention discloses a Deep Convolutional Network Aided Smart Agricultural Soil Enhancement Rover, an autonomous or semi-autonomous rover system (100) that integrates deep learning-based computer vision, edge computing, precision soil chemistry, and mechanical dispensing to combat desertification and enhance agricultural soil fertility. The rover is equipped with computer vision cameras that capture real-time images of sandy soil during traversal. These images are processed by a Convolutional Neural Network (CNN) model, preferably a ResNet architecture or alternatively MobileNet, VGG, YOLOv8, or EfficientNet, executing on an edge- mounted GPU, which classifies sand type and morphological characteristics including particle size, porosity, and angularity. Computer-controlled valves release the computed quantities into a blending chamber, producing ahomogeneous Liquid Nanoclay (LNC) and plant cellulose mixture, which is then dispensed through nozzle sprayers at the bottom of the rover. Fig. 1
Disclaimer: Curated by HT Syndication.
