|国家预印本平台
首页|Adaptive Vision-Guided Robotic Arm Control for Precision Pruning in Dynamic Orchard Environments

Adaptive Vision-Guided Robotic Arm Control for Precision Pruning in Dynamic Orchard Environments

Adaptive Vision-Guided Robotic Arm Control for Precision Pruning in Dynamic Orchard Environments

来源:Arxiv_logoArxiv
英文摘要

This study presents a vision-guided robotic control system for automated fruit tree pruning applications. Traditional agricultural practices rely on labor-intensive tasks and processes that lack scalability and efficiency, creating a pressing need for automation research to address growing demands for higher crop yields, scalable operations, and reduced manual labor. To this end, this paper proposes a novel algorithm for robust and automated fruit pruning in dense orchards. The proposed algorithm utilizes CoTracker, that is designed to track 2D feature points in video sequences with significant robustness and accuracy, while leveraging joint attention mechanisms to account for inter-point dependencies, enabling robust and precise tracking under challenging and sophisticated conditions. To validate the efficacy of CoTracker, a Universal Robots manipulator UR5e is employed in a Gazebo simulation environment mounted on ClearPath Robotics Warthog robot featuring an Intel RealSense D435 camera. The system achieved a 93% success rate in pruning trials and with an average end trajectory error of 0.23 mm. The vision controller demonstrated robust performance in handling occlusions and maintaining stable trajectories as the arm move towards the target point. The results validate the effectiveness of integrating vision-based tracking with kinematic control for precision agricultural tasks. Future work will focus on real-world implementation and the integration of 3D reconstruction techniques for enhanced adaptability in dynamic environments.

Dawood Ahmed、Basit Muhammad Imran、Martin Churuvija、Manoj Karkee

农业工程自动化技术、自动化技术设备计算技术、计算机技术

Dawood Ahmed,Basit Muhammad Imran,Martin Churuvija,Manoj Karkee.Adaptive Vision-Guided Robotic Arm Control for Precision Pruning in Dynamic Orchard Environments[EB/OL].(2025-04-09)[2025-05-18].https://arxiv.org/abs/2504.07309.点此复制

评论