Path Planning of Low-Altitude UAV for Tree Canopy Tracking and Orchard Monitoring

K. Kishore, Ritabrata Chakraborty
March 2025

Abstract

This paper presents a monocular vision-based approach for treeline following, designed to maintain stability under mild wind disturbances. In structured orchards, where trees are arranged in parallel rows, precise automated navigation is essential. Unlike traditional grid-based or path-trail methods, this work utilises instance segmentation to identify and track individual treelines, ensuring accurate differentiation. A curve-fitting algorithm generates a navigation line, with its offset and slope relative to the image centre guiding yaw and roll corrections, while pitch remains constant. The proposed method improves UAV stability and precision in orchard environments, advancing agricultural automation.

Type
Preprint
Publication
Manuscript in preparation

This research presents a monocular vision-based UAV system for autonomous treeline following in orchards. Using instance segmentation and curve-fitting algorithms, the system calculates yaw and roll corrections from the fitted line’s offset and slope, maintaining stability under wind disturbances.

UAV Navigation Monocular Vision Treeline Following Instance Segmentation Curve Fitting Wind Disturbance Orchard Automation Agricultural UAV computer-vision Precision Agriculture
Ritabrata Chakraborty
Ritabrata Chakraborty
CV Research Intern

Research Engineer specializing in robotics, computer vision, and autonomous systems. Currently developing automated data annotation solutions with foundation models for autonomous vehicles at Uber.