草莓采摘机器人小组

Paper Accepted by Computers and Electronics in Agriculture | 杨琳同学论文获《Computers and Electronics in Agriculture》接收

Fri, 18 Sep 2026 00:00:00 +0000

We are delighted to announce that our paper, “Harvesting Sequence Optimization for Clustered Strawberries Using a Fruit-Level Harvesting Difficulty Score,” has been accepted for publication in Computers and Electronics in Agriculture.

The paper is led by Lin Yang (杨琳), a Master’s student and the first author. The co-authors are Shimin Hu, Meili Sun, and Ya Xiong. This research was conducted by the Intelligent Equipment Research Center at the Beijing Academy of Agriculture and Forestry Sciences, in collaboration with the College of Information Science and Engineering at Shandong Agricultural University.

English

Harvesting strawberries that grow in dense clusters is challenging because individual fruits can differ substantially in accessibility, occlusion, and collision risk. Existing harvesting-sequence planning methods mainly aim to minimize manipulator travel distance, but often overlook the difficulty of harvesting each fruit.

This study proposes a difficulty-aware harvesting-sequence optimization framework. It combines:

a fruit-level Harvesting Difficulty Score (HDS) predicted by a two-stage convolutional neural network;
dynamic HDS updates after each virtual fruit removal; and
weighted multi-criteria sequence planning using genetic-algorithm-based (MOP-GA) and simulated-annealing-based (MOP-SA) solvers.

The HDS regression model achieved a mean absolute error of 6.55% and an R² of 72.39%. In offline experiments, MOP-GA reduced the average HDS per fruit to 0.282, compared with 0.291 for the travelling salesman problem (TSP) strategy and 0.297 for the bottom-up strategy.

Indoor robotic harvesting experiments across nine strawberry-cluster scenes further demonstrated lower HDS values for MOP-GA (0.197) and MOP-SA (0.204) than for TSP (0.235) and bottom-up (0.237). MOP-GA also achieved a lower observed harvesting failure rate than TSP (19.11% versus 26.67%), although the overall difference in failure probability among the methods was not statistically significant.

These results show that dynamically incorporating fruit-level harvesting difficulty can improve the feasibility of harvesting sequences in clustered environments. The work provides a promising foundation for safer and more reliable robotic strawberry harvesting, while real-time deployment will require further reductions in computational cost.

Congratulations to Lin Yang and all the co-authors on this achievement! 🍓🤖

中文

我们很高兴地宣布，实验室论文 《Harvesting Sequence Optimization for Clustered Strawberries Using a Fruit-Level Harvesting Difficulty Score》 已被国际期刊 《Computers and Electronics in Agriculture》 接收发表。

论文第一作者为硕士研究生 杨琳（Lin Yang），共同作者包括 Shimin Hu、Meili Sun 和 Ya Xiong。该研究由北京市农林科学院智能装备技术研究中心与山东农业大学信息科学与工程学院合作完成。

草莓在密集成簇生长时，不同果实在可接近性、遮挡程度和碰撞风险等方面存在明显差异，给机器人采摘带来很大挑战。现有采摘顺序规划方法通常侧重于缩短机械臂运动路径，而较少考虑单颗果实本身的采摘难度。

针对这一问题，本研究提出了一种面向采摘难度的草莓采摘顺序优化框架，主要包括：

利用两阶段卷积神经网络预测果实级 采摘难度评分（Harvesting Difficulty Score，HDS）；
每次虚拟移除果实后动态更新其余果实的 HDS；
采用基于遗传算法的 MOP-GA 和基于模拟退火的 MOP-SA，进行加权多目标采摘顺序规划。

HDS 回归模型的平均绝对误差为 6.55%，决定系数 R² 达到 72.39%。在离线实验中，MOP-GA 将单果平均 HDS 降至 0.282，优于旅行商问题（TSP）策略的 0.291 和自下而上策略的 0.297。

在 9 组草莓果簇场景的室内机器人实验中，MOP-GA 和 MOP-SA 的 HDS 分别为 0.197 和 0.204，均低于 TSP 的 0.235 和自下而上策略的 0.237。MOP-GA 的实际采摘失败率也低于 TSP（19.11% 对 26.67%），但不同方法之间总体失败概率的差异未达到统计显著水平。

研究结果表明，在采摘顺序规划中动态引入果实级采摘难度，可以提高机器人在密集果簇环境中的采摘可行性，为更加安全、可靠的草莓机器人采摘提供了新的思路。与此同时，算法计算成本仍是未来实现实时部署需要进一步解决的问题。

祝贺 杨琳同学 及全体作者！🍓🤖

Field Testing Begins for the WeedHitter 2 Laser Weeding Robot

Fri, 11 Sep 2026 00:00:00 +0000

We are excited to share that our team has begun field testing WeedHitter 2, the latest version of our laser weeding robot.

WeedHitter 2 features a new modular design that brings the platform closer to an industrial-ready system. The redesigned architecture supports more efficient integration, testing, maintenance, and future upgrades as we continue moving the technology from research prototypes toward practical agricultural deployment.

Alongside the development of the new robot, we have collected and annotated more than one million field data samples from farms across North China. This large-scale, real-world dataset captures diverse crops, weeds, growth stages, and field conditions, providing a strong foundation for training and evaluating the robot’s perception system.

Our current field trials are helping us assess the reliability and integration of the new modular platform under realistic operating conditions. The results will guide the next stage of development as we work toward a robust, scalable, and practical laser weeding solution for precision agriculture.

We look forward to sharing more progress from WeedHitter 2 soon. 🌱🤖

Two Reinforcement Learning Papers Accepted by IROS 2026

Mon, 06 Jul 2026 00:00:00 +0000

We are pleased to announce that two papers led by our master’s students, Changyou Miao and Li Teng, have been accepted for presentation at IEEE/RSJ IROS 2026.

Both works address the challenge of robotic harvesting in complex, occluded environments using reinforcement learning, with a shared focus on sim-to-real transfer and hierarchical policy design. The papers demonstrate how learning-based approaches can effectively handle the contact-rich, sequential nature of agricultural tasks such as obstacle separation and fruit detachment.

Changyou Miao (first author) and colleagues propose a reinforcement learning framework that unifies the full harvesting pipeline—obstacle separation, detachment, and placement—as a sequential decision-making problem. A hierarchical architecture combines high-level Cartesian actions with a low-level impedance controller for stable interaction under uncertain contact conditions. A feasibility-first observation alignment principle and domain randomization ensure robust zero-shot transfer from simulation to a structurally different real robot.

Li Teng (first author) and colleagues propose VGPA, a hierarchical reinforcement learning framework with a vision-guided decision mechanism and a Progressive Adaptive Exploration Strategy (PAES). The high-level vision module improves option selection and accelerates policy convergence, while PAES enhances exploration efficiency and training stability during continuous control. The framework is specifically designed for vision-based obstacle separation in clustered strawberry environments.

Congratulations to both teams on this achievement! 🎉

中文版

两篇强化学习论文被IROS 2026录用

我实验室两位硕士研究生 苗长友 和李腾作为第一作者的论文，近日被机器人领域顶级会议 IEEE/RSJ IROS 2026 正式录用。

两篇论文均聚焦于复杂遮挡环境下机器人采摘的挑战，采用深度强化学习方法，共同关注仿真到真实迁移与分层策略设计，展示了学习方法在处理农业操作任务中接触丰富、序列化决策问题上的有效性。

苗长友（第一作者）及合作者提出了一种强化学习框架，将避障、采摘和放置统一建模为序列决策问题。分层架构结合了高层笛卡尔空间动作与低层阻抗控制，确保在不确定接触条件下的稳定交互。可行性优先的观测对齐原则与域随机化策略，实现了从仿真到真实机器人的零样本迁移。

李腾（第一作者）及合作者提出了 VGPA 分层强化学习框架，包含视觉引导决策机制与渐进自适应探索策略（PAES）。高层视觉模块优化了选项选择并加速策略收敛，PAES则在连续控制学习中提升了探索效率与训练稳定性，专门面向簇生草莓环境下的视觉引导避障分离任务。

祝贺两位同学及合作团队！ 🎉

Vision-Based Fault Diagnosis and Self-Recovery for Strawberry Harvesting Robots Published

Mon, 06 Jul 2026 00:00:00 +0000

A paper led by PhD student Meili Sun has been published in Artificial Intelligence in Agriculture, titled “Vision-based early fault diagnosis and self-recovery for strawberry harvesting robots”.

The study presents an integrated vision framework that enables early fault diagnosis and self-recovery during robotic harvesting. Key innovations include:

SRR-Net: An end-to-end multi-task network for unified fruit and gripper perception.
Relative error compensation: Reduces picking point misalignment from 11.50 mm to 3.12 mm and 5.25 mm to 4.06 mm along the x- and y-axes, respectively.
Early abort strategy: Detects empty grasp/misgrasp early, saving approximately 0.5 s per failure case.
Slippage prediction & recovery: Achieved 88.89% prediction success and 81.25% recovery rate for slipping strawberries, saving 4.00 s per failure cycle.

A video demonstration is available at https://youtu.be/UOfwlHgXUgU.

DOI ↗

Congratulations to the authors! 🎉

我实验室博士生孙美丽为第一作者的论文 《基于视觉的草莓采摘机器人早期故障诊断与自恢复方法》 在 Artificial Intelligence in Agriculture 期刊正式发表。

该研究提出了一套集成视觉感知、故障诊断与自主恢复的完整框架，主要创新包括：

SRR-Net端到端多任务网络：统一感知果实与夹爪状态
相对误差补偿：将采摘点定位误差从11.50 mm和5.25 mm分别降至3.12 mm和4.06 mm
早期中止策略：提前检测空抓/误抓，每次失败节省约0.5 s
滑移预测与恢复：滑移预测成功率88.89%，恢复成功率81.25%，每次失败节省4.00 s

视频演示请见：https://youtu.be/UOfwlHgXUgU

DOI ↗

祝贺论文作者团队！ 🎉

Two Papers Accepted: Active Obstacle Separation and Real-Time Strawberry Grading

Tue, 24 Mar 2026 00:00:00 +0000

We are pleased to announce that two papers led by our master’s students have been accepted.

Quan Zhao (first author) and colleagues present a VLM-driven framework for active obstacle separation in robotic harvesting. The work achieved 89.4% clearing success in simulation and 83.9% in real-robot field trials. Accepted by Smart Agricultural Technology.

Jinshan Zhen (first author) and colleagues present a real-time RGB-D grading framework for strawberries under occlusions. The multi-attribute classifier achieved 93.65% overall accuracy with 29.7 FPS inference. Accepted by Smart Agricultural Technology.

Congratulations to both teams! 🎉

我实验室两位硕士研究生作为第一作者的论文近期被国际期刊 Smart Agricultural Technology 正式录用。

赵权（第一作者）及合作者提出一种基于视觉语言模型的主动清障框架，仿真环境清障成功率达89.4%，真实机器人试验达83.9%。

甄金山（第一作者）及合作者提出一种遮挡环境下草莓实时分级框架，多属性分类器整体准确率达93.65%，推理速度29.7 FPS。

祝贺两位同学及合作团队！ 🎉

Hybrid Robotic Gripper Paper Accepted by ICRA 2026

Mon, 02 Feb 2026 00:00:00 +0000

We are delighted to announce the acceptance of a paper to IEEE ICRA 2026, marking a remarkable achievement led by Xi Chen, an undergraduate intern from China Agricultural University. Co-first authored with Master’s student Yun Wang, their work on the “Hybrid Rigid-Soft Robotic Gripper with Shape Adaptation, Uniform Force Distribution, and Self-Locking Capabilities” will be presented in Vienna. This accomplishment highlights the exceptional potential of young researchers and our commitment to fostering talent through impactful projects.

Congratulations to Xi Chen, Yun Wang, and co-authors Lichao Yang, Haitao Li, and Ya Xiong!

我们非常高兴地宣布，一项由中国农业大学本科生实习生陈汐和硕士研究生汪赟为共同第一作者的研究成果被机器人领域顶级会议 IEEE ICRA 2026 录用，论文题为《一种具备形状适应、均匀力分布与自锁能力的刚柔混联机器人末端执行器》，将于2026年6月在维也纳进行报告。

本科生作为主要作者在顶级会议上发表论文实属难得，这充分展现了年轻研究者的卓越潜力和我实验室通过有影响力的项目培养人才的成果。

祝贺陈汐、汪赟以及合作者杨立超、李海涛、熊亚！

Lightweight Detector for Greenhouse Tomatoes Presented in New Paper

Mon, 22 Dec 2025 00:00:00 +0000

A paper led by master student Nengwei Yang, titled “PHDT-DETR: A Lightweight End-to-End Detector for On-Device Truss Tomato Detection in Greenhouses,” has been published in Smart Agricultural Technology.

The study proposes the PHDT-DETR, a streamlined detection model that maintains accuracy while being efficient enough to run in real-time on embedded devices within greenhouses, facilitating scalable robotic scouting and harvesting.

Congratulations to the authors!

New Multi-view Vision System Paper Accepted for Robust Strawberry Picking

Mon, 10 Nov 2025 00:00:00 +0000

A paper led by PhD student Shimin Hu, titled “Calibration-enhanced multi-view RGB-D vision for robust recognition and 3D localization of strawberries under occlusions,” has been accepted by Computers and Electronics in Agriculture.

The study presents a calibration-enhanced algorithm that fuses data from multiple cameras. This system significantly improves fruit recognition and precise 3D localization in dense, occluded environments, enabling more reliable robotic harvesting.

Congratulations to the authors!

Research on Light-Resilient Ripeness Detection for Strawberries Published

Tue, 28 Oct 2025 00:00:00 +0000

A paper led by PhD student Meili Sun, titled “Light-resilient visual regression of strawberry ripeness for robotic harvesting,” has been accepted for publication in Computers and Electronics in Agriculture.

The research introduces a robust visual regression model that maintains high accuracy in judging strawberry ripeness under varying greenhouse light conditions. This technology is key for enabling consistent, selective harvesting by robots.

Congratulations to the authors!

S-H Robotics Lab Co-organizes Agrirobotics Workshop at IROS 2025

Tue, 21 Oct 2025 00:00:00 +0000

On October 20, 2025, the IROS Workshop on Agricultural Robotics (Agrirobotics), co-organized by Dr. Ya Xiong alongside leading experts from Wageningen University, University of Lincoln, and Northwest A&F University, was successfully held at the top-tier robotics conference IROS.

The workshop focused on “Perception and Manipulation in Complex and Dynamic Agricultural Environments.” It featured keynote speeches, paper presentations, and panel discussions with participants from global institutions including Cambridge University, Carnegie Mellon University, UC Davis, and Zhejiang University. Awards for Best Scientific Contribution and Best Engineering Application were presented.

This event provided a vital platform for global innovation and collaboration in agricultural robotics, highlighting our lab’s active role in advancing the field.

Innovative Soft-Rigid Gripper Design Published in IEEE RA-L

Thu, 09 Oct 2025 00:00:00 +0000

A paper led by PhD student Lichao Yang, titled “High-Tolerance Soft-Rigid Gripper for Low-Damage Robotic Strawberry Harvesting,” has been published in the prestigious journal IEEE Robotics and Automation Letters (RA-L).

The paper presents a novel hybrid soft-rigid gripper design. Its structure adapts to fruit shape, tolerates positioning errors, and achieves very low damage rates, marking a significant advancement for reliable harvesting end-effectors.

Congratulations to the authors!

Paper on Strawberry Bruise Simulation Accepted by Postharvest Biology and Technology

Thu, 25 Sep 2025 00:00:00 +0000

A paper led by Zhicheng Hu, titled “Strawberry bruise susceptibility analysis based on discrete element method coupled with finite element simulation,” has been accepted for publication in Postharvest Biology and Technology.

This research combines Discrete Element Method (DEM) and Finite Element (FE) simulation to create a detailed model predicting bruise damage during robotic handling. The work provides crucial data for designing gentler robotic grippers to minimize fruit loss.

Congratulations to the authors!

We Showcase Full-Suite Agricultural Robots at 2025 Pepper Industry Summit

Wed, 17 Sep 2025 00:00:00 +0000

From September 15-16, 2025, the “2025 Pepper Industry (Jize) High-Quality Development Conference” was successfully held in Handan, Hebei Province. The conference served as a key platform for industry exchange, focusing on the standardization and brand development of the pepper industry.

During the event, attendees observed demonstrations of a full suite of agricultural robots independently developed by S-H Robotics, including models for laser weeding, inspection, pollination, and harvesting. The conference also marked the launch of a new collaborative R&D project between our lab and Tianye Company, aimed at developing a field transport robot.

Paper on Advanced Strawberry Pose Estimation Technique Published

Thu, 04 Sep 2025 00:00:00 +0000

A paper led by PhD student Tianxiao Zhu, titled “Five-degree-of-freedom strawberry pose estimation using key points and oriented bounding box detection,” has been published in Computers and Electronics in Agriculture.

This work enables 5-Degree-of-Freedom (5-DOF) pose estimation for strawberries by detecting key points and oriented boxes. Determining the fruit’s exact position and orientation is a critical step for planning damage-free robotic grasps.

Congratulations to the authors!

视频

Fri, 20 Jun 2025 00:00:00 +0000

通过视频了解我们的机器人研究、会议展示与田间试验。

草莓采摘机器人

田间试验 · 2024年5月 · 2:15

在商业温室中开展自主采摘，实现果实实时识别与轻柔抓取。

激光除草机器人 WeedHitter

精准农业 · 2025年3月 · 4:30

采用误差补偿轨迹与人工智能杂草识别的高精度激光除草系统。

定向授粉机器人

精准农业 · 2025年3月 · 4:30

多机械臂产生多方向气流振动，实现面向花朵的定向授粉。

Paper on Real-Time Strawberry Mass Estimation Accepted to IROS 2025

Sun, 15 Jun 2025 00:00:00 +0000

We are pleased to announce that the paper “Online Estimation of Table-Top Grown Strawberry Mass in Field Conditions with Occlusions”, led by Master’s student Jinshan Zhen, has been accepted for presentation at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025). This work addresses a critical challenge in precision agriculture: accurate fruit mass estimation under occlusion scenarios for robotics harvesting.

Technical Abstract

“Accurate mass estimation of table-top grown strawberries under field conditions remains challenging due to frequent occlusions and pose variations. This study proposes a vision-based pipeline integrating RGB-D sensing and deep learning to enable non-destructive, real-time mass estimation. The method employs YOLOv8-Seg for instance segmentation, CycleGAN for occluded region completion, and tilt-angle correction to refine frontal projection area calculations. A polynomial regression model maps geometric features to mass, demonstrating superior performance in occluded scenarios compared to traditional methods.”

Significance

This work enables:

Harvesting robot online grading through real-time mass feedback
Generalizable framework for other occluded fruit estimation tasks

Congratulations to Jinshan Zhen and the team! 🎉 We look forward to presenting at IROS 2025 in Hangzhou.

S-H Robotics Lab Showcases the Harvest-Flex Robot at SCO AI Forum 2025

Fri, 30 May 2025 00:00:00 +0000

On May 29, 2025, our “HarvestFlex-I” Strawberry Harvesting Robot was invited to exhibit at the China-Shanghai Cooperation Organization (SCO) Artificial Intelligence Cooperation Forum in Tianjin, under the theme “AI for China, AI for SCO”. The demonstration garnered high-level attention and media coverage.

Event Highlights

High-Profile Demonstration:
The robot performed live harvesting for distinguished guests including:
- Ms. Huang Ru, Member of the Leading Party Members Group, National Development and Reform Commission (NDRC)
- Mr. Li Shuqi, Member of the Tianjin Municipal People’s Government Leadership
- Mr. Azat Ibraimov, Director of the Presidential Executive Office, Kyrgyz Republic
Collaboration Opportunities:
Engaged with multiple potential partners on technology transfer and agricultural automation solutions.
Media Coverage:
Featured by CGTN and other international media outlets, highlighting China’s advancements in agricultural robotics.

Related Links:

Stay tuned for more updates from the S-H Robotics Lab! 🌱🤖

We Won Dual Awards at Xiong'an Smart Agriculture Competition

Wed, 28 May 2025 00:00:00 +0000

On May 27, 2025, our Laser Weeding Robot and Pollination Robot stood out among 100+ competing projects at the 2nd Xiong’an Future City Scenarios Competition - Smart Agriculture Track, both receiving Excellence Awards after rigorous evaluation.

Competition Journey

Showcase Stage: Live demonstrations of:
- WeedHitter laser weeding system’s 1.5 mm targeting precision
- BloomBot pollination robot’s 95% success rate in greenhouse trials
Evaluation Process:
- Preliminary review
- Semi-finals
- Final defense

HarvestFlex Strawberry Picking Robot Showcased at National Smart Agriculture Education Innovation Conference

Sun, 25 May 2025 00:00:00 +0000

On May 24, 2025, our HarvestFlex strawberry harvesting robot was invited to demonstrate at the National Smart Agriculture Education Innovation Conference in Wenzhou, Zhejiang, coinciding with the launch ceremony of the National Vocational College Smart Agriculture Skills Competition. The live picking demonstrations attracted widespread attention from agricultural educators and industry experts across China.

Event Highlights

Live Demonstrations:
Conducted multiple successful picking cycles, showcasing:
- Soft-gripping technology (<3% fruit damage rate)
- Real-time recognition (95% accuracy in clustered fruit conditions)
Expert Engagement:
Received optimization suggestions from several institutions on:
- Picking cycle time reduction
- End-effector design improvements
Industry-Academia Collaboration:
Initiated talks with many vocational colleges for:
- Training equipment purchasement

S-H Robotics Lab Members Attend IEEE RoboSoft 2025 in Switzerland

Mon, 28 Apr 2025 00:00:00 +0000

Dr. Xiong Ya and Dr. Lin Sen from the S-H Robotics Lab attended the 8th IEEE-RAS International Conference on Soft Robotics (RoboSoft 2025) in Switzerland from April 22-25, 2025, followed by an academic visit to EPFL (École Polytechnique Fédérale de Lausanne).

Conference Highlights

The conference focused on bio-inspired design, benchmarking, and industrial applications of soft robotics, attracting participants from over 20 countries. Key activities included:

Oral Presentation: Dr. Xiong Ya presented “High-Tolerance Soft-Rigid Gripper for Low-Damage Robotic Strawberry Harvesting”, showcasing collaborative advancements with the Haidian District Bureau of Agriculture.
Poster Session: Dr. Lin Sen shared research on selective harvesting robots and soft robotic end-effectors.
Competitions: The team observed competitions in raspberry harvesting and pipe inspection robotics, sparking discussions on agricultural applications.

Academic Visit to EPFL CREATE Lab

On April 23, the team visited the CREATE Lab at EPFL, exploring cutting-edge research including:

Low-cost tendon-driven robotic hands
Modular soft robotic arms
Transformable mobile platforms
Bio-inspired elephant trunk robots
Raspberry harvesting prototypes

草莓采摘机器人（HarvestFlex / 柔采）

Fri, 07 Mar 2025 00:00:00 +0000

项目概述

我们的草莓采摘机器人是一套面向草莓生产用工短缺和采摘效率问题的自主作业系统。机器人搭载先进的人工智能视觉系统和专利低损伤末端执行器，可精准识别并轻柔采摘成熟草莓。

演示视频： https://youtu.be/qBgaOyAfMuU

核心特点

人工智能视觉系统： 采用端到端草莓成熟度测量算法，克服光照变化，并通过多视角融合实现精准识别。
低损伤末端执行器： 专利夹持装置实现轻柔抓取，减少果实损伤。
LingXtend机械臂： 混联机械臂的工作空间约为UR3串联机械臂的3倍、IRB360并联机械臂的14倍，可适应温室避障需求。
在线分级： 采摘过程中按重量和成熟度在线分级，减少采后分选环节。
障碍物分离： 基于深度强化学习的障碍清除算法，提高密集果簇环境下的采摘性能。

应用价值

该机器人可减少对人工采摘的依赖、降低采收成本，并帮助保障草莓品质，尤其适用于面临用工短缺的大规模草莓生产场景。

返回研究项目

激光除草机器人（WeedHitter / 杂草打击者）

Thu, 06 Mar 2025 00:00:00 +0000

项目概述

我们的激光除草技术为传统化学除草和机械除草提供了高效替代方案。通过先进的误差补偿轨迹和优化参数组合，实现精准、高效的杂草清除。

核心特点

误差补偿轨迹： 支持水平、之字形、三角形和垂直等多种轨迹。
激光光斑调节： 光斑直径可在1–4毫米范围内调节。
优化作业参数： 采用最大激光功率、80°入射角和2毫米光斑直径等优化组合。

应用价值

该技术有助于减少对化学除草剂的依赖，推动可持续农业生产。

返回研究项目

Spraying Robots Modernize Greenhouse Farming in Ningjin County, Shandong province

Wed, 05 Mar 2025 00:00:00 +0000

In the modern agricultural innovation center of Baodian Town, Ningjin County, Shandong Province, our spraying robots are revolutionizing greenhouse farming. These robots use AI-powered navigation and high-precision sensors to monitor crops, apply water, nutrients, and pesticides with pinpoint accuracy, and ensure optimal growing conditions for tomatoes.

Impact on Agriculture

The robots help farmers reduce resource use, and address challenges like pests and nutrient deficiencies. This innovation highlights the growing role of smart agriculture in China.

Related Links:

Xinhua News Report

Stay tuned for more updates! 🌱🤖

番茄喷雾机器人

Wed, 05 Mar 2025 00:00:00 +0000

项目概述

我们的喷雾机器人利用人工智能和先进传感器，对水、营养液和农药进行精准施用，提升资源利用效率，减少浪费和环境影响。

核心特点

精准喷施： 面向目标区域按需施用水、营养液或农药。
自主导航： 在田间和温室内自主行驶、避障并优化路线。
实时监测： 利用传感器监测土壤水分、植株健康和环境状态。
灵活设置： 用户可通过移动端调整喷雾参数。

应用价值

该机器人有助于减少化学品用量、提高作物产量并推动可持续农业，适用于不同规模的农业生产场景。

返回研究项目

S-H Robotics Lab's Strawberry Harvesting Robot Showcased on CCTV-4

Tue, 04 Mar 2025 00:00:00 +0000

During the 2025 National Two Sessions, CCTV-4’s special program “Two Sessions I Care About” spotlighted cutting-edge agricultural technologies, featuring the strawberry harvesting robot developed by S-H Robotics Lab. The robot exemplifies China’s progress toward “high-quality development of agricultural machinery” as outlined in the 2025 Central Document No. 1.

Innovation Highlights

The fully autonomous robot, independently designed and patented by the S-H Robotics Lab, integrates AI and advanced engineering to address labor shortages and improve efficiency in strawberry farming. Some features include:

Rigid-Flexible Coupling Harvesting Gripper:
Reduces picking damage to 3%, matching manual harvesting precision, while offering superior error tolerance for varied fruit conditions.
LingXtend 1 Robotic Arm:
Performs complex motions like “splitting,” “bending,” obstacle avoidance, and mobile grasping, enabling dual-arm continuous harvesting in dense crop environments.
FieldRover Mobile Platform:
Achieves four-wheel differential steering, on-the-spot turning, diagonal walking, and precision speed control. Adjustable wheelbase and axle distance ensure adaptability to diverse field layouts.

Looking Ahead

As the lab continues to refine its designs, these advancements promise to empower farmers, reduce costs, and pave the way for a smarter, more sustainable agricultural future.

Related Links:

Stay tuned for more updates from the S-H Robotics Lab! 🌱🤖

温室番茄授粉机器人

Tue, 04 Mar 2025 00:00:00 +0000

项目概述

我们的温室番茄授粉机器人旨在解决温室环境中人工授粉效率低、劳动强度高等问题。机器人结合人工智能与精密机械技术，实现稳定、高效的授粉作业，助力提高产量并降低人工成本。

核心特点

智能授粉： 利用机器视觉识别花朵并实施精准授粉。
自主导航： 借助传感器和地图技术在温室内自主行驶。
轻柔高效： 面向娇嫩的番茄花设计，减少损伤并提高授粉成功率。

应用价值

该技术可降低对人工授粉的依赖，提高授粉一致性，助力温室番茄稳产增产。

返回研究项目

Laser Weeding Paper by Mr. Huayan Hu Accepted by IEEE Transactions on AgriFood Electronics

Thu, 20 Feb 2025 00:00:00 +0000

We are excited to announce that a paper by Mr. Huayan Hu, our master student, has been accepted by IEEE Transactions on AgriFood Electronics. Automatic targeting of weeds using lasers often encounters positional errors, particularly in dynamic weeding modes. The paper examines the positional error and weeding effect challenges in laser weeding, offering innovative solutions for precision and efficiency in laser weeding. Read the highlights below.

Highlights

Error Compensation Trajectories: Four weeding patterns were tested, with the horizontal trajectory proving most effective for reducing positional errors.
Laser Spot Adjustment: A device was designed to adjust spot diameter (1-4 mm), optimizing weeding efficiency.
Optimal Parameters: The best combination—horizontal pattern, maximum laser power, 80° incidence angle, and 2 mm spot diameter—achieved position errors under 2 mm.
Validation Results: Average cutting times for Chenopodium album, Polygonum hydropiper, Setaria viridis, and Eleusine indica were 0.411s, 0.308s, 0.419s, and 0.384s, respectively.

Congratulations to Mr. Huayan Hu on this outstanding achievement! 🌱🔬

S-H Robotics Lab Hosts Chinese New Year Seminar and Celebration

Mon, 20 Jan 2025 00:00:00 +0000

On January 20, 2025, the S-H Robotics Lab held its annual Chinese New Year Seminar, where lab members presented their progress in 2024, and outlined goals for the coming year.

Following the seminar, the lab celebrated the Lunar New Year with traditional activities, including paper cutting (剪纸), writing Spring Festival couplets (写春联), and decorating the lab (贴对联). The festivities created a joyful atmosphere, blending cultural traditions with team bonding.

The event concluded with a group dinner, where members enjoyed a feast and toasted to the lab’s successes and future opportunities. The celebration marked a strong start to 2025, filled with optimism and a shared commitment to excellence.

Paper Accepted in IEEE Transactions on AgriFood Electronics: A Fast Path-Planning Method for Continuous Harvesting of Table-Top Grown Strawberries

Thu, 09 Jan 2025 00:00:00 +0000

We are delighted to share that our paper, “A Fast Path-Planning Method for Continuous Harvesting of Table-Top Grown Strawberries,” has been accepted by the prestigious journal IEEE Transactions on AgriFood Electronics.

This work was primarily conducted by Mr. Yang Chen, our former Master’s student, with the conceptual foundation proposed by Dr. Ya Xiong.

Abstract Highlights:

Traditional collision-free path-planning algorithms, such as RRT and A*, often fall short in the context of efficient, continuous fruit harvesting due to their low search efficiency and excessive redundant points. To address this limitation, our research introduces the Interactive Local Minima Search Algorithm (ILMSA), a novel, fast path-planning method designed specifically for the continuous harvesting of table-top grown strawberries.

Key Features of ILMSA:

Interactive Node Expansion Strategy: Iteratively refines collision-free path segments using local minima points.
3D Path Optimization: Projects the 3D environment onto multiple 2D planes for optimal path generation, followed by smoothing 3D path segments.

Results:

Simulations:

ILMSA outperformed existing methods, achieving:

21.5% shorter path length and 97.1% faster planning time compared to 3D-RRT.
36.7% shorter paths and 97.8% faster planning time compared to the QAPF method in 3D environments.

Field Tests:

ILMSA demonstrated superior performance in real-world scenarios, achieving:

58% faster planning and execution times
69% shorter average path lengths compared to the LPS algorithm.

Conclusion:

This approach showcases the potential of ILMSA to enhance the efficiency and precision of continuous fruit harvesting.

We are proud of this achievement and extend our gratitude to all authors and collaborators involved in this research.

S-H Robotics Lab Successfully Completes 2024 Milestone Review of Two Agricultural Innovation Projects

Mon, 16 Dec 2024 00:00:00 +0000

Recently, the Haidiain District Bureau of Agriculture and Rural Affairs organized a milestone project completion review meeting for the 2024 Agricultural Science and Technology Innovation Projects. The meeting was attended by representatives from the district’s financial bureau, the Zhongguancun Science City Management Committee, and a panel of experts to assess the progress of projects hosted by S-H Robotics Lab.

Demonstration Highlights

The review panel visited two key demonstration sites to observe project outcomes:

Cuihu Smart Agriculture Innovation Factory (Phase II):
In the research greenhouse, the spraying robot and pollination robot efficiently navigated among lush tomato plants, autonomously performing tasks without any human intervention. Their flexible operations showcased the advancements in automation for greenhouse agriculture.
Zhongguancun Science Popularization Farm - “5G Strawberry Garden”:
The strawberry harvesting robot, equipped with a self-developed low-damage gripper and vision-based measurement algorithm, seamlessly moved under strawberry racks. The robot identified and picked ripe strawberries, offering a practical solution to seasonal labor shortages and rising costs in elevated strawberry farming.

Expert Feedback

The on-site demonstrations vividly showcased the robots’ capabilities, earning high praise from the expert panel for their practical applications. Following the field visit, the review team attended a detailed presentation covering project implementation, research achievements, and budget utilization. The experts engaged in a thorough discussion and unanimously approved the completion of the first phase of Haidian District’s Agricultural Science and Technology Innovation Projects.

Conclusion

The successful acceptance of these projects marks a significant milestone for S-H Robotics Lab. By addressing critical challenges such as labor shortages and operational efficiency, these projects represent an important step toward sustainable and intelligent farming solutions.

Nercita Leads National Key R&D Project on Smart Agricultural Equipment with Large Model Integration

Mon, 09 Dec 2024 00:00:00 +0000

We are thrilled to announce that the National Engineering Research Center for Information Technology in Agriculture (Nercita) has been approved to lead a major project under the 2024 National Key R&D Program. The project, titled “Application of Human-Machine-Environment-Agronomy Intelligent Collaborative Models in Unmanned Open-Field Vegetable Scenarios” (Project No. 2024YFD200800), represents a groundbreaking effort to integrate large-scale models into the field of smart agricultural equipment.

Project Highlights

Duration: December 2024 to November 2027
Total Funding: 21.5 million CNY, including 20 million CNY from central government grants
Project Leader: Dr. Ya Xiong, Nercita

Key Research Focus

Large Model Integration: Developing and applying advanced large-scale models to optimize collaborative interactions between humans, machines, and agricultural environments.
Unmanned Operations: Achieving full automation in large-scale open-field vegetable production.
Smart Equipment Development: Addressing complex challenges in task variability and labor-intensive operations through intelligent systems.

Participating Institutions

This national-level project involves collaboration with top-tier academic and research institutions, including:

China Agricultural University
Shandong Agricultural University
Yangzhou University
National Agricultural Mechanization Institute
Beijing Academy of Agriculture and Forestry Sciences

The approval of this project under the National Key R&D Program underscores the critical role of large models and cutting-edge technologies in advancing smart agricultural equipment. This initiative not only reflects Nercita’s leadership in agricultural robotics and AI integration but also marks a major step toward sustainable, intelligent, and efficient farming solutions.

Our Spraying Robots Recognized at the 2024 China Seed Congress

Fri, 18 Oct 2024 00:00:00 +0000

During the 2024 China Seed Congress held in Fengtai District Park from October 16th to 18th, representatives from the Fengtai District Government and the Beijing Municipal Bureau of Agriculture and Rural Affairs visited our exhibition area in the No.6 Intelligent Greenhouse. They observed our spraying robot in action and highly affirmed our innovative achievements in agricultural intelligence, commending the application prospects of our technology in smart agriculture.

Dr. Ya Xiong Presented at the "Smart Technology and Equipment Empowering High-Quality Development of the Vegetable Industry" Seminar in Beijing

Wed, 16 Oct 2024 00:00:00 +0000

On October 16-17, 2024, the “Smart Technology and Equipment Empowering High-Quality Development of the Vegetable Industry” seminar was successfully held in Beijing, under the guidance of the Beijing Association for Science and Technology and hosted by the Beijing Agricultural Association, Beijing Academy of Agriculture and Forestry Sciences, and Beijing Vegetable Association.

Dr. Ya Xiong was invited to participate in the event, where he delivered a report titled “Vegetable Harvesting Robots and Their Key Technologies.” A group of experts and scholars from the vegetable industry across China attended the conference to discuss cutting-edge innovations in smart technologies and equipment, seeking new paths for technology to promote high-quality agricultural development.

Welcoming Our New Team Members to S-H Robotics Lab!

Tue, 01 Oct 2024 00:00:00 +0000

We are excited to welcome three new members to the S-H Robotics Lab family: Mr. Yi Li, Mr. Donglin Li, and Ms. Guangying Cui. Their addition strengthens our team and enhances our capabilities.

Mr. Yi Li and Mr. Donglin Li: Both bring years of industry experience and will focus on the development and integration of software systems for agricultural robots. Their expertise will contribute to advancing our research and projects.
Ms. Guangying Cui: Joining as a project manager, she will take charge of laboratory and project management, ensuring smooth operations and efficient execution of our initiatives.

The addition of these members makes our team more complete.

S-H Robotics Lab Demoed the HarvestFlex Strawberry-Picking Robot to the Ministry of Agriculture Leadership

Tue, 10 Sep 2024 00:00:00 +0000

On September 10, 2024, Minister Han Jun and a delegation from the Ministry of Agriculture visited Nercita to examine the latest innovations in agricultural robotics. During the visit, the S-H Robotics team had the honor of demoing HarvestFlex, our newly developed strawberry-picking robot, to the visiting leadership.

This marked the first public unveiling of HarvestFlex, a robot designed and developed by our team for strawberry harvesting through advanced automation and precision technology. The visit provided an excellent opportunity to demonstrate our research and its potential to transform agricultural practices.

Showcasing Innovation - Field Rover, LingXtend and the spraying robot at the 2024 World Robotics Conference

Mon, 26 Aug 2024 00:00:00 +0000

We are excited to announce that our lab is showcasing three of our latest robotic innovations — the mobile platform ‘Field Rover,’ the robotic arm ‘LingXtend,’ and the spraying robot — at the 2024 World Robotics Conference. These agricultural robots have garnered significant attention from visitors across the globe, demonstrating the technology and forward-thinking design that our lab is proud to develop.

The ‘Field Rover’ and ‘LingXtend’ have not only captivated the conference attendees but have also been featured in numerous media outlets. Among the highlights, they were covered by Farmer’s Daily (in both print and app versions), China Agricultural Network, Beijing Daily App, Beijing Time, Beijing News, Haidian Newspaper, and the Beijing Haidian WeChat official account, among others. This widespread media coverage is a testament to the impact our work is having in the field of agricultural robotics.

In addition to the exhibition, Dr. Ya Xiong was invited to speak at the conference. His presentation focused on his work in agricultural robotics, addressing current challenges and sharing his insights on the future evolution of this rapidly advancing field. Dr. Xiong’s talk provided valuable perspectives on how agricultural robots are shaping the future of farming, emphasizing the importance of practical solutions and innovation in addressing the needs of modern agriculture.

联系我们

Sat, 24 Aug 2024 00:00:00 +0000

Dr. Shirin Ghatrehsamani from Penn State University Visited Nercita

Thu, 22 Aug 2024 00:00:00 +0000

On August 22, 2024, Dr. Shirin Ghatrehsamani, Assistant Professor in the Department of Agricultural and Biological Systems Engineering at Pennsylvania State University, visited Nercita. The visit was organized by Dr. Ya Xiong and included participation from Dr. Sen Lin, Dr. Ran Liu, Dr. Yueting Wang, along with more than 30 staff members and students from the center.

During her visit, Dr. Ghatrehsamani delivered a presentation on the work of the Precision Agriculture Research Group at Penn State University. She highlighted their research on robotic thinning, fruit harvesting, and the detection and treatment of crop diseases. Dr. Ghatrehsamani also provided insights into the future of robotic crop management, sparking in-depth discussions with Nercita researchers on key challenges such as the dynamic nature of orchard environments, obstacle handling, and the assessment of fruit damage.

Prior to the presentation, Dr. Ghatrehsamani was given a tour of the S-H Robotics laboratory, the Agricultural Robotics Laboratory, and the exhibition hall focused on agricultural information technology and intelligent equipment. These tours led to further discussions about the challenges in developing fruit and vegetable harvesting robots. Dr. Ghatrehsamani expressed regard for the progress Nercita has made in agricultural robotics and both parties agreed to pursue closer collaboration in future research and student training.

Dr. Ya Xiong Was Appointed as an Adjunct Professor and Graduate Supervisor at Shandong Agricultural University

Mon, 15 Jul 2024 00:00:00 +0000

On July 15, 2024, a significant event took place at the Beijing Academy of Agriculture and Forestry Sciences with the unveiling of the graduate internship base between the Information College of Shandong Agricultural University and Nercita. This collaboration marks a new chapter in academic partnership and mutual development.

During the ceremony, Dr. Ya Xiong was appointed as an adjunct professor and graduate supervisor at Shandong Agricultural University. The official letter of appointment was presented to Dr. Xiong by the dean of the college, Prof. Zhijun Wang.

In the next phase, both parties will continue to improve the collaborative mechanisms in place and work towards fostering meaningful and effective cooperation. Through this partnership, they hope to achieve outstanding outcomes in the joint training of graduate students, further enhancing research and educational excellence.

S-H Robotics Team Delivered Lectures to Undergraduate Students from Beijing University of Chemical Technology

Tue, 09 Jul 2024 00:00:00 +0000

On July 9, 2024, Nercita welcomed over 100 undergraduate students and faculty members from the School of Mechanical and Electrical Engineering at Beijing University of Chemical Technology for an on-site teaching session.

The students toured the Agricultural Information Technology and Intelligent Equipment Exhibition Hall, where Dr. Ya Xiong delivered a lecture titled “Overview of Machine Learning and Agricultural Robotics.” Dr. Xiong shared insights into the applications of machine learning in agricultural robotics, discussing both the opportunities and challenges within the field. PhD student Meili Sun followed with a lecture on “Agricultural Robots and Their Vision Systems.” Her presentation focused on the critical components of agricultural robots, including deep learning technologies and the role of vision systems in robotic harvesting.

The students were highly engaged throughout the visit, interacting with the S-H Robotics team to explore the forefront of smart agriculture and gain a deeper understanding of agricultural robotics and related technologies.

Yang Chen's Paper on LingXtend Accepted at IROS 2024

Sun, 30 Jun 2024 00:00:00 +0000

We are thrilled to announce that our master student Mr. Yang Chen’s paper, titled “Design and Control of a Novel Six-Degree-of-Freedom Hybrid Robotic Arm”, has been accepted for presentation at the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024). This paper introduces LingXtend, a novel hybrid six-degree-of-freedom (DoF) robotic arm, specifically designed to overcome the challenges faced by conventional robotic arms in agricultural settings.

Robotic arms are vital components of fruit-harvesting robots, yet traditional serial and parallel arms often struggle to meet the unique demands of agriculture. These include the need for a large workspace, rapid movement, effective obstacle avoidance, and affordability. LingXtend addresses these issues by combining the strengths of both parallel and serial mechanisms into a single hybrid arm.

Inspired by the flexibility and precision of yoga, LingXtend features two independently moving sliders along a single rail, functioning like feet. These sliders are connected through linkages and a meshed-gear set, enabling the arm to perform a “split” to maneuver under obstacles commonly found in greenhouses, such as pipes, tables, and beams. This innovative design allows the arm to maintain an optimal pose, even when mounted on a mobile platform, making it highly effective for dynamic fruit-picking tasks.

The hybrid arm also boasts a significantly larger workspace—nearly three times the volume of UR3 serial arms and fourteen times that of ABB IRB parallel arms.

Congratulations to Mr. Yang Chen on reaching this important milestone in his academic career!

Two New Projects Awarded Funding by Haidian District Bureau of Agriculture and Rural Affairs

Fri, 01 Mar 2024 00:00:00 +0000

We are pleased to announce that our lab has been awarded funding for two new projects by the Haidian District Bureau of Agriculture and Rural Affairs. These grants will support our ongoing efforts to advance agricultural robotics and integrate innovative technologies into modern farming practices.

The first project, titled “Integration of Agricultural Robotics and Agronomy for Hands-Free Strawberry Harvesting,” is led by Dr. Ya Xiong. This project focuses on integrating and demonstrating hands-free strawberry harvesting technology by combining agricultural robotics and agronomy. The project has received a grant of 2 million CNY for 2024, with a potential extension of an additional 4 million CNY over the next two years.

The second project, titled “Robotic Operation System for Protected Horticulture,” is led by Dr. Sen Lin. This project aims to develop a robotic operation system, including robotic spraying and pollination for greenhouse tomatoes. It has been granted 2.5 million CNY for 2024, with a possible extension of 2 million CNY in 2025.

These projects represent a significant step forward in our mission to revolutionize agriculture through the development and application of cutting-edge robotics. We are excited to embark on these initiatives and look forward to sharing our progress in the coming months.

实验室掠影

Mon, 24 Oct 2022 00:00:00 +0000

团队成员

Mon, 24 Oct 2022 00:00:00 +0000