Towards Robust UAV Navigation in Agriculture: Key Technologies, Application, and Future Directions

Area of Science:

• Agricultural Engineering
• Robotics
• Computer Vision

Background:

• Unmanned aerial vehicles (UAVs) are crucial for precision agriculture, enabling high-throughput sensing and field operations.
• Agricultural environments present unique navigation challenges like terrain variation and canopy occlusion, impacting UAV performance.

Purpose of the Study:

• To systematically review UAV navigation in agricultural settings from a system-level perspective.
• To identify challenges and propose future research directions for agricultural UAV navigation.

Main Methods:

• Summarized core technical components: sensing, localization, mapping, planning, and control.
• Analyzed navigation requirements across diverse agricultural scenarios (fields, orchards).
• Reviewed datasets, simulation platforms, and evaluation protocols.

Main Results:

• Identified key challenges: scene heterogeneity, perception degradation, task-semantic integration, control robustness, and lack of benchmarks.
• Highlighted varying navigation needs in open fields, orchards, and terraced farmland for applications like mapping and spraying.

Conclusions:

• Future research must focus on robust, task-aware, and modular navigation architectures.
• This will enable reliable and scalable deployment of UAVs in agriculture.