Environment simulator for studying automatic crop farming

Authors

  • Timo Oksanen Aalto University
  • M. Hakojärvi University of Helsinki, Department of Agricultural Sciences, Otaniementie 17, 02150 Espoo, Finland
  • T. Maksimow Aalto University, Department of Electrical Engineering and Automation, Finland
  • A. Aspiala Aalto University, Department of Electrical Engineering and Automation, Finland
  • M. Hautala University of Helsinki, Department of Agricultural Sciences, Otaniementie 17, 02150 Espoo, Finland
  • A. Visala Aalto University, Department of Electrical Engineering and Automation, Finland
  • J. Ahokas University of Helsinki, Department of Agricultural Sciences, Otaniementie 17, 02150 Espoo, Finland

Keywords:

robots, crop growth models, soil water models, precision agriculture, environment simulation, decision making, operation planning

Abstract

Agricultural machines capable of utilizing variable rate application technology are tackling spatial variability in agricultural fields.  Agricultural field robots are the next step in technology, robots which are capable of utilizing sensor and actuating technologies without human contact and operate only areas of interest.  However, agricultural field robots are still under research.  Robots are just one part of the next generation of crop farming having more advanced tools to do the work which currently requires humans.  The next generation of crop farming, in the vision of the authors, is based on automation, which incorporates stationary and moving sensors systems, robots, model based decision making, automated operation planning which adapts to spatial variability according to the measurements as well as to weather conditions.  This article presents a top-down approach of automated crop farming using simulation, trying to cover all the component parts on a fully automated farm.  In the article, the developed simulation platform is presented as well as sample simulation results.  The environment simulator is based on crop growth models, weed growth models, soil models, spatial variation generation and weather statistics.  Models for the environment were found in literature and were tailored and tuned to fit the simulation purposes, to form a collection of models.  The collection of models was evaluated by using sensitivity analysis.  Furthermore, a full scale scenario was simulated over one season, incorporating 9000 spatial cells in five fields of a farm.

 

Keywords: robots, crop growth models, soil water models, decision making, operation planning

Downloads

Published

2014-03-28

Issue

Section

V-Management, Ergonomics and Systems Engineering