Robotic control of broad-leaved dock

Broad-leaved dock ( Rumex obtusifolius L.) is a common and troublesome grassland weed with a wide geographic distribution. In organic far ming, the best option to control the weed is manual removal of the plants. In this report we des cribe the development and first tests of a robot to detect and control broad-leaved dock. An a nalysis of requirements led to the construction of a diesel-powered frame of 1.25 x 1. 11 m to which four independently driven wheels are attached. Weeds are detected with a downward-looking camera that provides fullcolour images with a resolution of 1.5 mm per pixel . Image processing is based on Fourier analysis of sub-images (tiles) of 8x8 pixels. Weeds are controlled using the method proposed by Austrian farmer F. Riesenhuber. This method consists of a chopper with a single 0.20 m blade that rotates around a vertical axis at 1500 r pm and is pushed into the ground at the location of the weed. In field tests the robot was run at 0.5 m/s. Under favourable conditions, more than 90% of weeds were detected and positioning of the chopper occurred with adequate precision. The time required to position a nd operate the chopper was determined to be 12 s. Approx. 25% of controlled weeds exhibited regrowth. We conclude that our robot provides an attractive alternative to manual remova l of broad-leaved dock.

[1]  J. V. Stafford,et al.  Robotic agriculture - the future of agricultural mechanisation? , 2005 .

[2]  John E. E. Ebinger.  Rumex obtusifolius L. , 1979 .

[3]  W. Bond,et al.  The biology and non-chemical control of broad-leaved dock (Rumex obtusifolius L.) and curled dock (R. crispus L.) , 2007 .

[4]  W. Kühbauch,et al.  A new algorithm for automatic Rumex obtusifolius detection in digital images using colour and texture features and the influence of image resolution , 2007, Precision Agriculture.

[5]  E. Rogers.  Diffusion of Innovations , 1962 .

[6]  G. van Straten,et al.  Autonomous Navigation with a Weeding Robot , 2006 .

[7]  Gerrit Polder,et al.  Weed Detection Using Textural Image Analysis , 2007 .

[8]  S. Vougioukas,et al.  A two-stage optimal motion planner for autonomous agricultural vehicles , 2006, Precision Agriculture.

[9]  Herwart Böhm,et al.  Überprüfung der Effektivität der maschinellen Ampferregulierung im Grünland mittels WUZI unter differenzierten Standortbedingungen , 2004 .

[10]  E. Noe,et al.  Problems with disseminating information on disease control in wheat and barley to farmers , 2008, European Journal of Plant Pathology.

[11]  Urs Niggli,et al.  Effects of nitrogen fertilization and cutting frequency on the competitive ability and the regrowth capacity of Rumex obtusifolius L. in several grass swards , 1993 .

[12]  J. Harper,et al.  Rumex Obtusifolius L. and R. Crispus L. , 1964 .

[13]  Noboru Noguchi,et al.  Agricultural automatic guidance research in North America , 2000 .

[14]  N. D. Tillett,et al.  DEALING WITH COLOR CHANGES CAUSED BY NATURAL ILLUMINATION IN OUTDOOR MACHINE VISION , 2004, Cybern. Syst..

[15]  Rob Cramb,et al.  Processes Influencing the Successful Adoption of New Technologies by Smallholders , 2000 .

[16]  A. Schut,et al.  Imaging spectroscopy for early detection of nitrogen deficiency in grass swards , 2003 .

[17]  J. Schellberg,et al.  Identification of broad-leaved dock (Rumex obtusifolius L.) on grassland by means of digital image processing , 2006, Precision Agriculture.

[18]  R. Haggar,et al.  The effects of Rumex obtusifolius on the seasonal yield of two mainly perennial ryegrass swards. , 1983 .