Aarhus University Seal / Aarhus Universitets segl

Peter Ahrendt

Trends in Robotic Sensor Technologies for Fruit Harvesting: 2010-2015

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

DOI

  • Andrejs Zujevs, Vidseme University of Applied Sciences, Letland
  • Vitalijs Osadcuks, Mitavas roboti Ltd, Letland
  • Peter Ahrendt
In the modern world processes and technologies tend to be automated, autonomous and precise. The world population is constantly growing and thus food production technologies should be brought to a qualitatively new level. Quality requirements for food products also tend to increase and become more complex. Precision Agriculture provides the possibility to use soil more intelligently and effectively. Precision Agriculture includes sensor technologies for yield mapping and measuring, soil sensing, nutrient and pesticide application, irrigation control, robotic harvesting, etc. With the increase of the density and energy effectiveness of computing power, it has also become possible to use open source libraries to incorporate complex signal processing, object detection and machine learning into embedded applications. These factors have led to a situation where designs of commercially successful robotic plant inspection and harvesting solutions can emerge. This paper provides a review of modern sensor systems used in semi or fully automated robotic harvesting, including fruit detection and localization prior to pick or slice. Sensors used in selective harvesting were also reviewed. Sensor systems were classified in the following categories: computer vision, chemical sensors, tactile sensors and proximity sensors. The main trends in the future of robotic harvesting will involve usage of combinations of different sensor systems that provide accuracy and reliability.
OriginalsprogEngelsk
TidsskriftProcedia Computer Science
Vol/bind77
Sider (fra-til)227-233
Antal sider7
ISSN1877-0509
DOI
StatusUdgivet - 31 dec. 2015

Se relationer på Aarhus Universitet Citationsformater

ID: 96793210