Solar Power Based Multipurpose Agriculture Robot with Leaf-Disease Detection

Abstract

Real-time collision free route tracking presents increasingly difficult challenges for mobile robots operating in big, dynamic areas. Finding a practical collision-free path for the robot to go along while in a cultivated area is thus a crucial necessity. In India, agriculture is a major source of revenue. Bacterial and fungal illnesses are brought on by conditions including diseases, insect infestations, and abrupt climatic changes. Early illness detection and taking certain steps will produce better outcomes. The Agritech robot manoeuvres in a dynamic environment to look for tomato plant illnesses. While travelling around an agricultural area, the robot takes pictures of the plants and stores them in a database for comparison and analysis. These pictures are sorted as either sick or well. When a plant is infected, the pesticide sprayer unit will activate and spray the plant with pesticide. A tool is being created to detect three prevalent ailments that affect tomato plants, including Bacterial Blight, Leaf Spot, and Anthracnose. The identification of diseases is carried out using a trained neural network model. The Convolution Neural Network, K-means clustering, and Support Vector Machine utilised for image segmentation and illness classification, respectively, are trained into the model. The employment of autonomous robots in tomato plantations can lower the danger of human mistake associated with manual spraying, which makes them very beneficial in agriculture.