Information about the Sub-Projects
Development and test of a real-time process control for a sensor-based fertilizer application system
Site specific application of nitrogen fertilizer by the means of a real-time sensor-based system with map overlay (e.g. of yield maps) requires an advanced process control with support of expert systems. For open agricultural electronic communication systems this could be implemented by an specific "In-field Controller". Following four topics have to be considered: Takeover of tasks including sensor-orientated set points and yield mapping information of several years, sensor data acquisition and processing, derivation of site specific application set points by appropriate expert rules and documentation of site specific information.
The integration of the "In-field Controller" into variable rate fertilization systems with a task controller and an application controller enables such a sophisticated process control and offers possibility to adapt and scale existing application systems to future sensor technologies.
Within this partial project such a sensor-based process control with map overlay will be specified for intensive nitrogen fertilization and implemented and tested in a lab environment. System architecture is based on an Agricultural BUS-System compliant to ISOBUS / LBS (ISO 11783 / DIN 9684).
To realize this aim two main objectives have to be achieved:
Optimisation of Site Specific Crop Management with Contact-Free Sensors regarding to crop situation
||Maidl, Dr. Franz - Xaver
||+49 (0)8161 71-3425
||+49 (0)8161 71-3488
Agricultural crops often show a high site specific variability regarding growth and yield. Based on the knowledge of how yield growing processes are turning at heterogen laocations, the sub-project 09 is going to develop a site specific crop management system. This will consider the the site specific soil potential as well as the actual nutrition situation of the crops. This way a method for a variable and crop adjusted nitrogen fertilization will be established.
Destructive methods so far for analyzing soil and plants are not suitable for site specific applications because of the high number of samples, that would be needed. Therefore non-destructive online-methods like spectral measurement are going to developed further on.
The Investigations will take place in georeferenced trial field plots on heterogen fields of the experimental farms of TU Munich.
- Estimation of biomass and N-uptake in earlier stages of development of winter wheat and mais
- Generating vegetation indices fpr estimation of biomass and N-uptake of barley and potatoes
- Minimizing negative influences on spectral measurement under field conditions especially at earlier stages of crop development
- Testing this non-destructive method in winter wheat and mais cultivation using and further developing the algorithms from the IKB I project
- Developing algorithms for site specific crop management for barley and potatoes
Scale Dependent Validation and Compare of Contact-Free Sensors at Site Specific Crop Farming over several years
On large cultivation units taking care of the site specific variability of soil quality seems to be more and more important. With GPS and non-destructive measurement it is possible to difference bi plots into several smaller management units. A central point after generating management units is the adjusted nitrogen fertilization of crops within these units, because time- and volume-adjusted nitrogen nutrition is very important to realize the highest posssible yield on a site. Fpr measurement of nitrogen status of crops a field spectrometer is used. Testing the relation between spectral measurement under field conditions and the real plant nutrition state of the crops is the aim of the sub-project.
For the trial there will be generated a articicial variability within a homogen area. There will be situated trial plots with different nitrogen nutrition niveaus and different seed volumes. For the investigations different cultivars of winter wheat and mais will be used. The spectral measurement will be realized with the 'Hydro-Agri N-Sensor'. At the same time a part of the plants from the trial plots will be harvested and analyzed by classical destructive methods for nitrogen contents and dry matter. Reaktime sensor data and ground truth data will be fit together by Geographic Information Sytems (GIS).
Deriving of Decision Support Rules for Site Specific N-Fertilization Based on Database-aided Information und Expert Knowledge
The aim of sub-project 12 is to derive decision rules for site specific farming, which can be used for automated processing in field. The base are extensive site specific data collected by mobile processing systems in field. This data contain process data information as well as sensor data with information of soil and crop variability. The main points of interests are information about soil conductivity, spectral information of crop reflection behavier and draft force measurement information from tillage. Those data, which are stored in a central project database, will be analyzed by using methods of data mining. The data mining should offer information for optimized application volumes and strategies at nitrogen fertilizing.
The extracted rules will be validated by agronomists and tested by farm trials.
Economical and Ecological Vallidation of Site Specific Crop Farming
Sub-project 13 is going to deliver an extended analysis of economical and ecological aspects of different approaches of site specific crop farming on the field and on the farm level. Especially the time-related variability of crop farming processes should be considered at the investigations to show the chances as well as the limits of site specific variable rate fertilisation.
Another topic is to investigate, how site specific fertilizing strategies could help to minimize negative economical cosequences at crop farming under ecological restrictions (like a maximum tolerable nitrogen wash out).
For both, ressource use and economical aspects in conjunction with site specific crop farming, a book of rules should be written.
Melody Thomas, 06.04.2004, Imprint