Search

Aim of study: An approach to integrate knowledge into the IT-infrastructure of precision agriculture (PA) is presented. The creation of operation relevant information is analyzed and explored to be processed by standardized web services and thereby to integrate external knowledge into PA. The target is to make knowledge integrable into any software solution. Area of study: The data sampling took place at the Heidfeld Hof Research Station in Stuttgart, Germany. Material and methods: This study follows the information science’s idea to separate the process from data sampling into the final actuation through four steps: data, information, knowledge, and wisdom. The process from the data acquisition, over a professional data treatment to the actual application is analyzed by methods modelled in the Unified Modelling Language (UML) for two use-cases. It was further applied for a low altitude sensor in a PA operation; a data sampling by UAV represents the starting point. Main results: For the implemented solution, the Web Processing Service (WPS) of the Open Geospatial Consortium (OGC) is proposed. This approach reflects the idea of a function as a service (FaaS), in order to develop a demand-driven and extensible solution for irregularly used functionalities. PA benefits, as on-farm processes are season oriented and a FaaS reflects the farm’s variable demands over time by origin and extends the concept to offer external know-how for the integration into specific processes. Research highlights: The standardized implementation of knowledge into PA software products helps to generate additional benefits for PA.

In the given study an adjustable harrowing system is presented and tested. The automatic harrow can increase or decrease the harrowing intensity during operation. A gentle (20%), medium (40%) and aggressive (60%) harrow intensity was chosen. Prior to the application, measurements were performed in each plot concerning the weed density and composition. With this information, a Decision Support System based on fuzzy logic was used in order to trigger an appropriate tine angle movement. Thus, areas with high crop and weed densities were applied with more aggressive harrowing treatments and areas with lower weed densities with a gentler treatment. A spring barley field was adopted to evaluate the suitability and effectiveness of the system at the University of Hohenheim, Germany. A harrow application was conducted at the maximum permitted harrow intensity (60%), as the farmer would have applied it on the field and an automatic adaptation, based on the results of the decision support system. A further herbicide treatment and an untreated control were also included. Weed counting was performed prior to and after the application, along with biomass cuts and yield in order to estimate the treatment efficacy. The automatic system performed well, providing similar results as the nonautomatic harrowing, but with lower intensity levels. Dicotyledonous weeds were, in both mechanical applications, reduced as well as by the herbicide application, without any significant differences.

The weed suppressive ability of oilseed radish (Raphanus sativus var. oleiformis Pers.) cover crop is attributed to high competitiveness for resources and biochemical effects on weeds. The oilseed radish cover crop was sown in five treatments plus an untreated control over a period of five weeks before and three weeks after winter wheat harvest. Additionally, fertilization effects on oilseed radish biomass and weed suppression were measured. The highest biomass of the cover crop was observed 12 weeks after harvest (WAH) when the oilseed radish was sown one week after harvest (1 WAH) (2015) and five weeks before harvest (5 WBH) (2016). No differences of fertilization were observed concerning oilseed radish and weed biomass in 2015, whereby increased biomass was found after fertilization in 2016. The highest weed control efficacy of up to 83% and 90% was achieved in treatments 1 WAH (2015) and 5 WBH (2016) at 12 WAH. The early sowing of oilseed radish in winter wheat resulted in low germination and biomass yield within the field, due to low precipitation in 2015. Nevertheless, there is a high potential of early sown oilseed radish for higher weed control efficacy, which was demonstrated in 2016.

A non-destructive measuring technique was applied to test major vine geometric traits on measurements collected by a contactless sensor. Three-dimensional optical sensors have evolved over the past decade, and these advancements may be useful in improving phenomics technologies for other crops, such as woody perennials. Red, green and blue-depth (RGB-D) cameras, namely Microsoft Kinect, have a significant influence on recent computer vision and robotics research. In this experiment an adaptable mobile platform was used for the acquisition of depth images for the non-destructive assessment of branch volume (pruning weight) and related to grape yield in vineyard crops. Vineyard yield prediction provides useful insights about the anticipated yield to the winegrower, guiding strategic decisions to accomplish optimal quantity and efficiency, and supporting the winegrower with decision-making. A Kinect v2 system on-board to an on-ground electric vehicle was capable of producing precise 3D point clouds of vine rows under six different management cropping systems. The generated models demonstrated strong consistency between 3D images and vine structures from the actual physical parameters when average values were calculated. Correlations of Kinect branch volume with pruning weight (dry biomass) resulted in high coefficients of determination (R2 = 0.80). In the study of vineyard yield correlations, the measured volume was found to have a good power law relationship (R2 = 0.87). However due to low capability of most depth cameras to properly build 3-D shapes of small details the results for each treatment when calculated separately were not consistent. Nonetheless, Kinect v2 has a tremendous potential as a 3D sensor in agricultural applications for proximal sensing operations, benefiting from its high frame rate, low price in comparison with other depth cameras, and high robustness.

Non-chemical weed control methods need to be directed towards a site-specific weeding approach, in order to be able to compete the conventional herbicide equivalents. A system for online weed control was developed. It automatically adjusts the tine angle of a harrow and creates different levels of intensity: from gentle to aggressive. Two experimental plots in a maize field were harrowed with two consecutive passes. The plots presented from low to high weed infestation levels. Discriminant capabilities of an ultrasonic sensor were used to determine the crop and weed variability of the field. A controlling unit used ultrasonic readings to adjust the tine angle, producing an appropriate harrowing intensity. Thus, areas with high crop and weed densities were more aggressively harrowed, while areas with lower densities were cultivated with a gentler treatment; areas with very low densities or without weeds were not treated. Although the weed development was relatively advanced and the soil surface was hard, the weed control achieved by the system reached an average of 51% (20%–91%), without causing significant crop damage as a result of harrowing. This system is proposed as a relatively low cost, online, and real-time automatic harrow that improves the weed control efficacy, reduces energy consumption, and avoids the usage of herbicide.