English

智能化农业装备学报(中英文) ›› 2025, Vol. 6 ›› Issue (2): 58-68.DOI: 10.12398/j.issn.2096-7217.2025.02.005

• • 上一篇    下一篇

传感器技术在水果采摘机器人中的应用现状及趋势

吕秋辉1(), 朱立学1,2(), 张世昂1, 陈逸鹏1, 毛顺1   

  1. 1.仲恺农业工程学院机电工程学院,广东 广州,510225
    2.农业农村部华南果蔬绿色防控重点实验室,广东 广州,510225
  • 收稿日期:2025-03-13 修回日期:2025-04-18 出版日期:2025-05-15 发布日期:2025-05-20
  • 通讯作者: 朱立学
  • 作者简介:吕秋辉,男,2001年生,广东高州人,硕士研究生;研究方向为农业智能技术。E-mail: 137637519@qq.com
  • 基金资助:
    国家自然科学基金项目(32101632);广州市科技计划项目(2023B03J0862)

Application status and trends of sensor technology in fruit picking robots

LÜ Qiuhui1(), ZHU Lixue1,2(), ZHANG Shi'ang1, CHEN Yipeng1, MAO Shun1   

  1. 1.College of Mechanical and Electrical Engineering,Zhongkai Universiy of Agriculture and Engineering,Guangzhou 510225,China
    2.Key Laboratory of Green Prevention and Control on Fruits and Vegetables in South China,Guangzhou 510225,China
  • Received:2025-03-13 Revised:2025-04-18 Online:2025-05-15 Published:2025-05-20
  • Contact: ZHU Lixue
  • About author:LÜ Qiuhui, E-mail: 137637519@qq.com
  • Supported by:
    National Natural Science Foundation of China(32101632);Guangzhou Science and Technology Program Project(2023B03J0862)

摘要:

目前水果采摘存在劳动力短缺、采摘效率低和作业环境复杂等问题,亟须发展具备高精度感知与自主作业能力的智能化采摘装备,以全面提升果实采摘的效率和质量。传感器技术在水果采摘机器人中的应用包括路径规划、果实识别、定位及抓取控制等关键任务环节。针对非结构化果园环境,视觉、触觉与激光传感器的协同应用可实现目标识别、位置感知与避障控制,显著提升了采摘机器人对复杂环境的适应能力与作业精度,但是现有传感器仍然存在一些技术短板,如视觉传感器易受阳光干扰、枝叶遮挡和果实密集分布等因素影响,导致目标检测困难;触觉传感器易受温湿度影响,难以量化复杂的力学反馈,因而细微抓取力控制困难;由于非结构化环境下的路径优化困难,且激光传感器成本高昂,限制了其大规模应用。同时,单一传感器存在感知维度单一、环境适应性不足和果实特征感知不足等局限,难以适应非结构化果园环境。为此,针对多传感器融合技术面临的数据异构性、时序同步性和计算复杂性等挑战,对传感器技术在水果采摘机器人的未来应用进行了展望,指出融合红外、紫外等多波段成像技术和高动态范围(high dynamic range imaging,HDR)成像技术,柔性电子皮肤结合仿生结构设计的多传感器融合技术有望得到广泛应用。

关键词: 传感器技术, 视觉传感器, 触觉传感器, 激光传感器, 多传感器融合技术, 农业机器人

Abstract:

Contemporary fruit harvesting operations are confronted with several challenges, including labor shortages, low picking efficiency, and complex operational environments. These factors underscore the imperative for developing intelligent harvesting equipment endowed with high-precision perception and autonomous operation capabilities to comprehensively enhance the efficiency and quality of fruit picking. Sensor technology plays a critical role in key tasks of fruit-picking robots, such as path planning, fruit recognition, localization, and grasping control. In unstructured orchard environments, the collaborative utilization of vision, tactile, and laser sensors facilitates target identification, positional awareness, and obstacle avoidance, significantly improving the robot's adaptability and operational accuracy in complex settings. However, existing sensors still have technical limitations. For example, visual sensors are susceptible to deleterious effects from ambient sunlight interference, occlusion from branches and leaves, and the dense clustering of fruits, all of which impede robust target detection. Tactile sensors often display sensitive to fluctuations in temperature and humidity, which poses challenges for the accurate quantification of complex mechanical feedback, thereby hindering precise control of gripping force. In addition, path optimization in unstructured environments remains challenge, and the high procurement cost of laser sensors limits their large-scale application. Single-sensor systems suffer from limitations such as single-dimensional perception, poor adaptability to environmental variations, and insufficient recognition of fruit characteristics, making them suboptimal for deployment in unstructured orchard conditions. Therefore, this paper explores the future applications of sensor technologies in fruit-picking robots, focusing on the challenges of multi-sensor fusion, including data heterogeneity, temporal synchronization, and computational complexity. It highlights that the integration of multi-spectral imaging technologies such as infrared and ultraviolet, high dynamic range (HDR) imaging, and flexible electronic skin combined with biomimetic structural designs in multi-sensor fusion systems holds great promise for widespread application.

Key words: sensor technology, visual sensors, tactile sensors, laser sensors, multi-sensor fusion technology, agricultural robotics

中图分类号: