Robotic Weeding (2001-2006)

Vision and Objectives

The vision of the project is to contribute to the development of novel weeding technologies that can reduce manual effort by 50-100% in organically grown sugar beets and vegetables and herbicide usage by 75-100% in high value crops, e.g. conventional grown sugar beets. An important aim is to establish a synergistic framework for key-scientists, the commercial partners, PhD and MSc-students carrying out research on robotic weeding. The specific objectives of the project within this framework are

• to optimise (technically and biologically) the geo-referencing of seeds during the sowing operation
• to develop automated identification and geo-referencing of crop and weed seedlings
• to investigate the accuracy, potentials and limitations of non-chemical devices for intra-row and close-to-crop weeding
• to investigate targeted micro-spraying of weeds in the close-to-crop area

The goal of the project is to demonstrate an integration of crop-seed mapping, computer vision for identification crop and weed seedlings and highly accurate devices for mechanical and chemical targeting of weed seedlings. The investigations and demonstrations will be carried out in fields with sugar beets but the detection and weeding principles studied in the project can be modified to work with any wider spaced crops (densities < 20 plants m-2), e.g. maize and vegetables.

 

Funding

This project is part of the National Research programme ‘Sustainable Technology in Agriculture’ ("Bæredygtigt teknologi i Jordbruget”), supported by the Danish Technical Research Council (STVF), the Danish Agricultural and Veterinary Research Council (SJVF) and the Danish Ministry of Food, Agriculture and Fisheries.

Duration

2003 - 2005

Project partners

Project leader:

Svend Christensen, Research Director at DIAS Research Center, Bygholm

Other contributors:
Hans W. Griepentrog, Associate Professor at KVL, Frederiksberg

Michael Nørremark, PhD student at KVL, Frederiksberg

Henning Tangen Søgaard, Senior Researcher at DIAS Research Center, Bygholm

Ivar Lund, PhD student at DIAS Research Center, Bygholm

Enrico Graglia, Researcher, DIAS, Flakkebjerg

Organisation of the project

• WP1: System definition and integration (DIAS + KVL)
  - HAKO (KVL)
  - API (DIAS)
• WP2: Seed geo-referencing (KVL)
• WP3: Computer vision for identification of crop seedlings (DIAS)
• WP4: Non-chemical weeding tools (KVL)
  - Close-to-crop weeding
  - GPS controlled cycloid hoe
  - GPS and Side-shift controlled Inter-row Hoeing
• WP5: Micro-spray system (DIAS)

Publications

 

Cooperation

• Eco-Dan A/S, Bøgeskovvej 6, DK-3490 Kvistgaard
• Hardi International A/S, Helgeshøj Alle 38, DK-2630 Taastrup
• Osnabrück University of Applied Sciences, Postfach 19 40, D-49009 Osnabrück
• Hako-Werke GmbH&Co, Hamburger Str. 209-239, D-23843 Bad Oldesloe