Detection and classification of plant pots in real time using artificial intelligence methods for mobile manipulators used in nursery farms and greenhouses

利用人工智能方法对苗圃和温室中使用的移动机械手进行花盆实时检测和分类

• 批准号: 538450-2019
• 负责人: Najjaran, Homayoun
• 金额: $ 1.82万
• 依托单位: University of British Columbia
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2019
• 资助国家: 加拿大
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=680447
• 关键词: Detection; classification; plant; pots; real

Advanced Intelligent Systems (AIS), a Canadian company located in Burnaby, BC, Canada is a new, but growing, company conducting world-class R&D on Unmanned Systems with applications in farming and agriculture. At AIS the R&D activities are directed towards the use of unmanned systems in nursery farms and greenhouses for moving planted pots using a mobile manipulator designed and manufactured AIS to expand the use cases of the AIS proprietary platform. This research project focuses on the design and development of algorithmic procedures to enable the existing mobile manipulator of AIS called BigTop (i.e., a wheeled mobile base equipped with an articulated robotic arm on top) to identify, track, classify planted pots, pick up and deliver them to their desired locations using onboard sensors, namely cameras, IMUs, LiDAR, and wheel encoders, fully autonomously and without any human intervention. The applicant and his research team will provide the expertise required for i) computer vision methods to recognize and estimate the pose of planted pots for proper grasping, and ii) the simultaneous localization and mapping (SLAM) methods for autonomous operation of the mobile platform.

先进智能系统公司(AIS)是一家位于加拿大卑诗省本纳比的加拿大公司，是一家新兴但正在成长中的公司，致力于在农业和农业应用方面进行世界级的无人系统研发。在AIS，研发活动的目标是在苗圃农场和温室中使用无人系统来移动种植的花盆，使用设计和制造的移动机械手来扩大AIS专有平台的使用案例。本研究项目专注于设计和开发算法程序，使AIS现有的移动机械手Bigtop(即顶部配备关节机械臂的轮式移动基座)能够使用车载传感器，即摄像头、IMU、LiDAR和车轮编码器，完全自主地识别、跟踪、分类种植的花盆，并将它们运送到期望的位置，而无需任何人为干预。申请人和他的研究团队将提供所需的专业知识，i)计算机视觉方法，以识别和估计种植的盆栽的姿势，以便适当地抓取，以及ii)同步定位和测绘(SLAM)方法，用于移动平台的自主操作。