Bewertung der Praxisnähe und des Lösungsbeitrags von  Technologien in der Landwirtschaft 4.0 am Beispiel digitaler  Höhenmodelle, drohnengestützter Pflanzenbonitur und mechanisch-chemischer Unkrautregulierung

Pommerehne, Luisa

by Luisa Pommerehne

Doctoral thesis

Date of Examination:2024-02-28

Date of issue:2024-12-19

Advisor:Prof. Dr. Frank Beneke

Referee:Prof. Dr. Frank Beneke

Referee:Prof. Dr. Achim Spiller

Referee:Prof. Dr. Anne-Katrin Mahlein

Persistent Address: http://dx.doi.org/10.53846/goediss-10966

Files in this item

Name:+ + + DISSERTATION LP ABGABEVERSION OHNE MAR...pdf

Size:2.90Mb

Format:PDF

ViewOpen

The following license files are associated with this item:

Abstract

English

Digital technologies are omnipresent and open up new opportunities for agriculture. At the same time, challenges arise when implementing the technologies in practice. OUTCOME SITUATION: Over the years, significant advances have been made in digital agriculture to increase the efficiency and sustainability of crop production. These advances are based on technological innovations in data collection, processing and analysis in conjunction with sensors, robotics and information and communication technologies. Despite the early stage of development of digital agriculture, the technical transformation is proceeding at a rapid pace due to the increasing diversity and complexity of agricultural machinery and process chains. The challenge lies in meeting the needs of individual farms and utilising the benefits of innovations in a way that is adapted to the heterogeneous conditions of agricultural use. PROBLEM AND TASK: The integration of new, digital technologies into agricultural practice has been slow. Technology manufacturers are faced with the problem of evaluating the potential of relevant technologies, taking into account the individual and heterogeneous framework conditions in agriculture, in order to optimise their own technology and implement it in practice. The task is to develop an easy-to-use, customer-orientated and easy-to-transfer procedure that enables agricultural technology manufacturers to carry out a technology comparison among rapidly changing, current technologies in agriculture. RELEVANCE: Against the backdrop of global food security and environmental protection, increased efforts must be made to ensure this. Agriculture 4.0 plays a central role here, as new digital technologies promise to improve sustainability, productivity and efficiency in agricultural production. A customer-oriented technology comparison by manufacturers can help to optimise product development, strengthen market adaptability and thus contribute to the provision of pesticide-saving and efficient technologies for agricultural practice. SOLUTION AND METHODOLOGICAL APPROACH: By combining and categorising different parameters, a systematic comparison method for technologies is developed that can be easily transferred to other agricultural use cases. The conceptual basis is provided by a literature analysis on the introduction of technologies in agriculture as well as the core elements "perceived usefulness" and "perceived ease of use" of the Davis technology acceptance model. These are transferred to the manufacturer's perspective and specified for practical application. The developed technology acceptance assessment procedure (TABV) is tested on the basis of three current case studies from agricultural practice. The current use cases are as follows: I: generation of point clouds, II: digital plant scoring and III: weed control with hoeing machine and band sprayer. To this end, field trials will be set up in three use cases and the following specific questions will be investigated using different methodological approaches: 1) To what extent do the technologies to be compared differ in their technological contribution to the solution? 2) To what extent do the technologies to be compared differ in their practical relevance from the farmer's point of view, taking into account user costs, user-friendliness, user usability and application time requirements? RESULTS: The TABV is made up of parameter A, "technological solution contribution", and parameter B, "practical relevance". The technological solution contribution is defined as a relative performance advantage in a group comparison and indicates the extent to which the functionality of the technology serves the application purpose. Practical relevance is defined as a measure of operational applicability and practical feasibility. Integration into the agricultural operation is made possible by taking into account user costs, ease of use, seasonal/location dependent user usability and application time requirements according to a tier system. Research question 1, to what extent the technologies to be compared differ in their technological solution contribution, is answered as follows for the following three current use cases. The solution contribution "external accuracy" in use case I indicates how close the point clouds generated by different technologies are to an independent reference point cloud. A comparison is made between ground-based and airborne platforms that are equipped with different sensor systems and are located at different recording heights. These are compared with a reference measurement by calculating the standard deviations (SD) of the Euclidean distances (<= 1 m) to the reference point cloud: The lowest SD of 0.038 m was measured by the combination of aircraft and LiDAR scanner, followed by the UAV laser scanner variant (0.084 m). The platform furthest from the ground, the satellite, achieved an SD of 0.57 m in the target/actual point cloud evaluation. The solution contributions of use case II are represented by the detection and counting of weeds and crops. A comparison is made between drone-based scoring and manual scoring in two field trials. Methodologically, the scoring types are compared using mean differences and Spearman's rank correlation. Both sites show negative mean differences due to a tendency towards higher crop coverage (KDG) and plants per square metre in the manual scoring. For KDG, 76 % of the mean values of the manual data collection at site 1 and 62 % at site 2 show higher values compared to the UAV-based scoring, in relation to the total differences. The Spearman correlation coefficient shows a medium to strong positive linear relationship between the two scoring methods for all sites and parameters. This effect is more pronounced for the parameters relating to crop growth. The KDG shows a strong correlation between UAV and manual scoring, r = 0.998 for site 1 and Spearman r = 0.994 for site 2. There is also a strong positive linear correlation for the crop plants per square metre, with r = 0.943 for site 1 and r = 0.917 for site 2. The technologies compared in use case III contribute to almost complete weed control without negative effects on the crop plant. The comparison is made between the hoeing machine with band sprayer (HB), conventional surface application and the combination of both technologies for post-emergence weed control in the sugar beet crop at two locations. Using a mixed model, several contrast tests were carried out to compare the different levels of variants. At the last observation time, the following could be determined: The mean differences between the variants three times surface spraying and two times use of the hoeing machine band sprayer were statistically significant for the parameters KDG, crop size, weed size and weeds per m2. The largest values for KDG, crop size and crops per m2 were found in the area spraying, followed by the combination variant, while the HB had the lowest values in the variant comparison. Apart from the parameter K/m2, these trends were also confirmed at location 1. The weed size and the number of weeds per m2 increased in the comparison of variants in the order of surface application to combination variant to HB. This was also confirmed at site 1. At both sites, higher efficiencies were achieved with conventional surface application (92.7 % and 99.5 %) compared to HB (87 % and 98 %). Research question 2, to what extent the technologies to be compared differ in their practical relevance, is answered as follows for the following three current use cases. According to the TABV, in use case I, the agricultural tractor equipped with GNSS RTK is the technology with the greatest practical relevance due to its simple handling. As this technology is part of the tractor's guidance system, there are no additional costs. The technology can be used repeatedly in day-to-day work with little training at any location and in almost all weather conditions. If GNSS RTK technology is not available, point clouds for DEMs from satellites and aeroplanes should be used as an alternative for practical reasons. This data service is constantly available and does not require any field work. However, its use is seasonal and locally limited. Surveying with the three technologies tractor, satellite and aircraft involves the least additional work for the user. The other LiDAR-equipped technologies - UAV, car and robot - require the most additional time. In use case II, the intensity of manual scoring by hand is associated with a significantly higher expenditure of time (4 min/m2) and costs (14,333 €/ha) compared to drone flights by a service provider (0.008 min/m2; 1,610 €/ha). The use of drones requires legal and technical knowledge of drone control and software application. Manual scoring is less demanding in terms of location and time in everyday use and can therefore be used more flexibly. In use case III, the pure HB variant in contract work proves to be the most cost effective choice (€248.67 net) with a 50% saving on pesticides and a high level of user friendliness. The combination variant and the self-mechanised field sprayer have comparable costs (€349.86 net and €397.27 net respectively), but place different demands on the operating conditions: The use of a hoeing machine and band sprayer in a combined process places higher demands on the operating conditions. The working time requirement of the self-mechanised trailed 27 m wide crop protection sprayer is low at 0.39 ha/h. RESULT OF CRITICAL ASSESSMENT: The TABV method offers an initial coverage of the complex adaptation and diffusion behaviour of agricultural users when new technologies are introduced. Limitations in the application result from the specific focus on one target group and the limited transferability to other technology fields or target groups, as the method is based on a sector-specific literature analysis. OUTLOOK: To summarise, a technology acceptance assessment procedure (TABV) was developed in this work. It is based on a broad literature analysis. As a result, the TABV consists of the parameters practical relevance and technological solution contribution and enables agricultural machinery manufacturers to make a customer-oriented comparison of technologies. The procedure was successfully applied to three crop cultivation case studies. Due to the trend towards digitalisation, the TABV will gain in importance in the future, supported by the potential of new technologies, which can also be seen in the applications mentioned above (e.g. UAV inspections have the potential to reduce costs by a factor of 9). The approach of the developed TABV is transparent, easy to communicate and can be used by agricultural machinery manufacturers, but also by decision-makers in politics and industry as well as in further research projects. Further research is needed to ensure the validation of the approaches developed in this work through multi-year field trials at various locations and to test and further develop the TABV in other applications.

Keywords: digital technologies; digital technologies; digital technologies

Statistik