Smart Agricultural Sprayer System using Weighted KNN on Autonomous Farming Drones

Authors

  • Arjon Turnip Universitas Padjadjaran, Indonesia Author
  • Mohammad Taufik Universitas Padjadjaran, Indonesia Author
  • Poltak Sihombing Universitas Sumatera Utara, Indonesia Author
  • Endra Joelianto Institut Teknologi Bandung, Indonesia Author
  • Noman Naseer Air University, Islamabad, Pakistan Author

Keywords:

Smart agriculture, Autonomous farming drones, K-Nearest Neighbors, Sprayer system, Wireless

Abstract

Precision agriculture has gained significant attention for its potential to optimize resource use and enhance crop productivity. Autonomous drones equipped with smart systems are increasingly being used for tasks such as watering, fertilization, and monitoring. This paper presents the development and evaluation of a Smart Sprayer system that utilizes an autonomous drone for precise water discharge control. The system employs a weighted K-Nearest Neighbors (KNN) algorithm to classify watering intensity, achieving an accuracy of 90.7%. A dataset of 3,750 samples was used, evenly distributed across six classes (Stop, Very Low, Low, Medium, High, and Very High) and split 80:20 for training and testing. The model demonstrated strong performance in key metrics such as Precision and Recall, excelling particularly in predicting features like Altitude and Sprayer Valve, as reflected in high Positive Predictive Values and low False Discovery Rates. However, challenges remain in accurately classifying certain features, including Drone Velocity and GPS Error, where lower Precision and higher False Discovery Rates were observed. These limitations highlight the need for further model refinement and tuning. Future work will focus on real-world deployment and hardware optimizations to enhance system performance in autonomous agricultural operations.

Downloads

Published

2024-12-01

How to Cite

Smart Agricultural Sprayer System using Weighted KNN on Autonomous Farming Drones. (2024). Internetworking Indonesia Journal, 16(2), 29-35. https://internetworkingindonesia.org/index.php/iij/article/view/53