UAV Path Planning Model Leveraging Machine Learning and Swarm Intelligence for Smart Agriculture

Abstract

Smart agriculture, through precision farming, is revolutionizing traditional farming methods by optimizing resource use and enhancing yields. With the integration of technology, especially the advent of Unmanned Aerial Vehicles (UAVs) or drones, modern agriculture has attained new heights in efficient crop management, real-time data collection, and sustainable practices. UAVs play a pivotal role, offering aerial insights into crop health, soil conditions, and targeted resource application, promoting sustainable farming. However, navigating UAVs efficiently across dynamic agricultural terrains presents challenges, particularly in path planning. While traditional grid-based models have their merits, the complexities of modern farms demand more adaptive models. This work introduces a hierarchical path planning framework for UAVs, combining the “Enhanced Genetic Algorithm using Fuzzy Logic” for global planning and the “Improved D* Algorithm” for real-time local adjustments. This dual-layered approach ensures efficient, safe, and energy-conserving UAV trajectories, marking a significant advancement in UAV-based smart agriculture.