Evaluating performance and generalizability of Learning from Demonstration for the harvesting of apples & pears DOI Creative Commons
Robert van de Ven, Ard Nieuwenhuizen,

E.J. van Henten

и другие.

Smart Agricultural Technology, Год журнала: 2025, Номер unknown, С. 101006 - 101006

Опубликована: Май 1, 2025

Язык: Английский

Design, integration, and field evaluation of a robotic blossom thinning system for tree fruit crops DOI
Uddhav Bhattarai, Qin Zhang, Manoj Karkee

и другие.

Journal of Field Robotics, Год журнала: 2024, Номер 41(5), С. 1366 - 1385

Опубликована: Март 28, 2024

Abstract The United States (US) apple industry relies heavily on semi‐skilled manual labor force for essential field operations such as training, pruning, blossom and green fruitlet thinning, harvesting. Blossom thinning is one of the crucial crop‐load management practices to achieve desired crop load, fruit quality, return bloom. While several techniques chemical mechanical are available large‐scale approaches often yield unpredictable results may damage canopy, spurs, leaf tissue. Hence, growers still depend laborious, labor‐intensive, expensive hand outcomes. This research presents a robotic solution precision in orchards using deep learning‐based computer vision system, six‐degrees‐of‐freedom UR5e manipulator, an electrically actuated miniature end‐effector. integrated system was evaluated commercial orchard which showed promising targeted selective thinning. Two approaches, center boundary were investigated evaluate system's ability remove varying proportions flowers from flower clusters. During end‐effector around cluster boundary, while involved actuation only at centroid fixed duration 2 s. Field evaluation that approach thinned 67.2% clusters with cycle time 9.0 s per cluster, whereas 59.4% 7.2 cluster. Upon further improvement adoption, proposed could help address problems faced by current hand, chemical, approaches.

Язык: Английский

Процитировано

10

Enhanced Deep Learning Model for Apple Detection, Localization, and Counting in Complex Orchards for Robotic Arm–Based Harvesting DOI Creative Commons

Tantan Jin,

Xiongzhe Han, Pingan Wang

и другие.

Smart Agricultural Technology, Год журнала: 2025, Номер 10, С. 100784 - 100784

Опубликована: Янв. 11, 2025

Язык: Английский

Процитировано

2

Dynamic Task Planning for Multi-Arm Apple-Harvesting Robots Using LSTM-PPO Reinforcement Learning Algorithm DOI Creative Commons
Zhengwei Guo,

Heng Fu,

Jiahao Wu

и другие.

Agriculture, Год журнала: 2025, Номер 15(6), С. 588 - 588

Опубликована: Март 10, 2025

This paper presents a dynamic task planning approach for multi-arm apple-picking robots based on deep reinforcement learning (DRL) framework incorporating Long Short-Term Memory (LSTM) networks and Proximal Policy Optimization (PPO). In the context of rising labor costs shortages in agriculture, automated apple harvesting is becoming increasingly important. The proposed algorithm addresses key challenges such as efficient coordination, optimal picking sequences, real-time decision-making complex, orchard environments. system’s performance validated through simulations both static environments, with demonstrating significant improvements completion time robot efficiency compared to existing strategies. results show that LSTM-PPO outperforms other methods, offering enhanced adaptability, fault tolerance, execution efficiency, particularly under changing unpredictable conditions. research lays foundation development more efficient, adaptable robotic systems agricultural applications.

Язык: Английский

Процитировано

2

Developments in Automated Harvesting Equipment for the Apple in the Orchard: Review DOI Creative Commons

Yi Tianjing,

Mustafa Mhamed

Smart Agricultural Technology, Год журнала: 2024, Номер 9, С. 100491 - 100491

Опубликована: Дек. 1, 2024

Harvesting apples is one of the most apple-challenging operations; its process labor-intensive, and for various reasons, automation has yet to advance as swiftly it might. Researchers have concentrated on developments in robotics automated apple harvesting, two domains with a plethora opportunities difficulties that require more evaluation future growth quality. In this paper, we provide an overview harvesting by beginning perspective focuses vision techniques recognition systems. We then cover outcomes, methods, time, observations via robust analysis, including visible light, spectral, thermal imaging. After that, were followed localization apple, which aids detaching from branches, leaves, other overlapping apples, besides directing end-effectors grip remove apples. Next, harvester robots progress includes machinery equipment contain grippers, arms, manipulators, speed up operations upgrade performance. Additionally, platforms aid boost productivity, reduce demand strength, lower danger accidents at work. Furthermore, discussion part comprehensive analysis covering works detection systems robot technology. Finally, summarize challenges, limitations, opportunities, perspectives trends technologies. This research offers several avenues advancement interaction fields attract investment firms, such sorting bagging. Assist sustaining expansion communities offering services raise yield quality fruit.

Язык: Английский

Процитировано

9

High-precision fruit localization using active laser-camera scanning: Robust laser line extraction for 2D-3D transformation DOI Creative Commons
Pengyu Chu, Zhaojian Li, Kaixiang Zhang

и другие.

Smart Agricultural Technology, Год журнала: 2024, Номер 7, С. 100391 - 100391

Опубликована: Янв. 4, 2024

Recent advancements in deep learning-based approaches have led to remarkable progress fruit detection, enabling robust identification complex environments. However, much less has been made on 3D localization, which is equally crucial for robotic harvesting. Complex shape/orientation, clustering, varying lighting conditions, and occlusions by leaves branches greatly restricted existing sensors from achieving accurate localization the natural orchard environment. In this paper, we report design of a novel technique, called Active Laser-Camera Scanning (ALACS), achieve localization. The ALACS hardware setup comprises red line laser, an RGB color camera, linear motion slide, external RGB-D camera. Leveraging principles dynamic-targeting laser-triangulation, enables precise transformation projected 2D laser surface apples positions. To facilitate pattern acquisitions, Laser Line Extraction (LLE) method proposed high-precision feature extraction apples. Comprehensive evaluations LLE demonstrated its ability extract patterns under variable occlusion conditions. system achieved average apple accuracies 6.9 - 11.2 mm at distances ranging 1.0 m 1.6 m, compared 21.5 commercial RealSense indoor experiment. Orchard that 95% detachment rate versus 71% By overcoming challenges research contributes advancement harvesting technology.

Язык: Английский

Процитировано

7

Application of Soft Grippers in the Field of Agricultural Harvesting: A Review DOI Creative Commons
Daode Zhang, Wei Zhang, Hualin Yang

и другие.

Machines, Год журнала: 2025, Номер 13(1), С. 55 - 55

Опубликована: Янв. 14, 2025

This review summarizes the important properties required for applying soft grippers to agricultural harvesting, focusing on their actuation methods and structural types. The purpose of is address challenges limited load capacity stiffness, which significantly hinder broader application in agriculture. paper examines research progress variable stiffness over past five years. We categorize various techniques analyze advantages disadvantages enhancing capacity, dexterity, degree integration, responsiveness, energy consumption grippers. applicability limitations these context harvesting are also discussed. concludes that combined material technology with a motor claw structure better suited operations woody crops (e.g., apples, citrus) herbaceous tomatoes, cucumbers) unstructured environments.

Язык: Английский

Процитировано

1

Active Laser-Camera Scanning for High-Precision Fruit Localization in Robotic Harvesting: System Design and Calibration DOI Creative Commons
Kaixiang Zhang, Pengyu Chu, Kyle Lammers

и другие.

Horticulturae, Год журнала: 2023, Номер 10(1), С. 40 - 40

Опубликована: Дек. 31, 2023

Robust and effective fruit detection localization is essential for robotic harvesting systems. While extensive research efforts have been devoted to improving detection, less emphasis has placed on the aspect, which a crucial yet challenging task due limited depth accuracy from existing sensor measurements in natural orchard environment with variable lighting conditions foliage/branch occlusions. In this paper, we present system design calibration of an Active LAser-Camera Scanner (ALACS), novel perception module robust high-precision localization. The hardware ALACS mainly consists red line laser, RGB camera, linear motion slide, are seamlessly integrated into active scanning scheme where dynamic-targeting laser-triangulation principle employed. A high-fidelity extrinsic model developed pair laser illumination enabling precise computation when target captured by both sensors. random sample consensus-based then designed calibrate parameters based collected data. Comprehensive evaluations conducted validate scheme. results show that proposed method can detect remove data outliers achieve parameter computation, calibrated able maximum measurement error being than 4 mm at distance ranging 0.6 1.2 m.

Язык: Английский

Процитировано

11

Autonomous multiple‐trolley collection system with nonholonomic robots: Design, control, and implementation DOI
Peijia Xie, Bingyi Xia,

Anjun Hu

и другие.

Journal of Field Robotics, Год журнала: 2024, Номер unknown

Опубликована: Июль 18, 2024

Abstract The task of collecting and transporting luggage trolleys in airports, characterized by its complexity within dynamic public environments, presents both an ongoing challenge a promising opportunity for automated service robots. Previous research has primarily developed on universal platforms with robot arms or focused handling single trolley, creating gap providing cost‐effective efficient solutions practical scenarios. In this paper, we propose low‐cost mobile manipulation incorporated autonomy framework the collection transportation multiple that can significantly enhance operational efficiency. method involves novel design mechanical system vision‐based control strategy. We lightweight manipulator docking mechanism, optimized sequential stacking trolleys. On basis Control Lyapunov Function Barrier Function, controller online Quadratic Programming, which improves accuracy. application our is demonstrated real‐world scenarios, where it successfully executes multiple‐trolley task.

Язык: Английский

Процитировано

4

In-Depth Evaluation of Automated Fruit Harvesting in Unstructured Environment for Improved Robot Design DOI Creative Commons
Sadaf Zeeshan,

Tauseef Aized,

Fahid Riaz

и другие.

Machines, Год журнала: 2024, Номер 12(3), С. 151 - 151

Опубликована: Фев. 22, 2024

Using modern machines like robots comes with its set of challenges when encountered unstructured scenarios occlusion, shadows, poor illumination, and other environmental factors. Hence, it is essential to consider these factors while designing harvesting robots. Fruit are automatic that have the ability improve productivity replace labor for repetitive laborious tasks. Therefore, aim this paper design an improved orange-harvesting robot a real-time environment orchards, mainly focusing on efficiency in occlusion varying illumination. The article distinguishes itself not only efficient structural but also use enhanced convolutional neural network, methodologically designed fine-tuned dataset tailored oranges integrated position visual servoing control system. Enhanced motion planning uses rapidly exploring random tree star algorithm ensures optimized path every activity. Moreover, proposed machine rigorously tested validate performance fruit robot. unique aspect in-depth evaluation test five areas include accurate detection fruit, time picking, success rate damage picked as well consistency picking illumination occlusion. results then analyzed compared previous study most aspects environment.

Язык: Английский

Процитировано

3

Robotic Multi-Boll Cotton Harvester System Integration and Performance Evaluation DOI Creative Commons
Shekhar Thapa, Glen C. Rains,

Wesley Porter

и другие.

AgriEngineering, Год журнала: 2024, Номер 6(1), С. 803 - 822

Опубликована: Март 13, 2024

Several studies on robotic cotton harvesters have designed their end-effectors and harvesting algorithms based the approach of a single boll at time. These systems often slow times per due to limited computational speed extended time taken by actuators retract for picking individual bolls. This study modified design previous version end-effector with aim improving ratio boll. fabricated pullback reel pull plants backward while rover harvested moved down row. Additionally, YOLOv4 detection model hierarchical agglomerative clustering algorithm were implemented detect bolls cluster them. A was then developed harvest in clusters. The end-effector, reel, vacuum conveying system, model, algorithm, straight-line path planning integrated into small red rover, both lab field tests conducted. In tests, robot achieved 57.1% an average 2.5 s 56.0%, it took 3.0 Although there no improvement setting over design, robot’s performance significantly better, 16% higher 46% reduction compared tested 2022.

Язык: Английский

Процитировано

3