UAV-Based Coverage Path Planning for Unmanned Agricultural Vehicles.

Area of Science:

• Agricultural Engineering
• Robotics
• Geospatial Analysis

Background:

• Autonomous navigation in agriculture requires accurate path planning.
• Manual map creation for agricultural vehicles is labor-intensive and prone to errors.

Purpose of the Study:

• To develop an Unmanned Aerial Vehicle (UAV)-based coverage path planning system for automating agricultural vehicle guidance.
• To reduce operator intervention in creating navigation maps and improve operational efficiency.

Main Methods:

• Utilized low-altitude UAV photogrammetry to generate high-resolution orthophoto maps for spatial information.
• Developed automatic generation of travel and working paths from operator-selected anchor points in image coordinates.
• Implemented Mercator projection for converting pixel coordinates to geographic coordinates (latitude and longitude).
• Created a Graphical User Interface (GUI) for path generation, visualization, and evaluation.

Main Results:

• Path planning calculation accuracy showed a mean planar coordinate error of 2.23 cm and a maximum error of 3.37 cm.
• Field tests with an unmanned high-clearance sprayer demonstrated lateral navigation errors within ±5.5 cm.
• The system proved feasible and highly accurate for autonomous agricultural vehicle navigation.

Conclusions:

• The developed UAV-based coverage path planning method offers a feasible and accurate solution for autonomous agricultural vehicle operations.
• This technology effectively automates guidance and reduces operator workload in precision agriculture.
• It represents a significant advancement towards fully autonomous agricultural vehicle systems.