Browsing by Person "Griepentrog, Hans"

Now showing 1 - 6 of 6

Aspects of incorporating biodegradable textiles to improve sports turf
(2023) Stürmer-Stephan, Bastian; Griepentrog, Hans
Due to climate change and the need to save water, water consumption must be reduced not only in agriculture but also in urban areas. There are 55,072 sports fields in Germany that have to be irrigated in summer. In order to reduce the amount of irrigation, two approaches were researched and discussed in this thesis. The first approach is to adapt new sports fields to the local weather conditions. This approach is a decision support system, based on a model. The input variables are recorded weather data from the German Weather Service for the location where the new sports field is to be built, the hydrological properties of the substrates, and the expected costs. An optimized dimensioning of the rootzone layer is calculated by an EA solver of Microsoft Excel. This thickness of the layer can be used for the construction project. This was calculated exemplary for 3 locations. The presented model needs to be further evaluated through field trials. For existing sports fields, the root zone layer can only be changed with great effort. In this case, a biodegradable nonwoven can be installed in an existing sports field with drainage layer structure. This nonwoven transport water from the deeper drainage layer into the root zone of the turf through the capillaries, so that the water is available to the turf. To achieve this function, the 150 mm wide nonwoven must be installed vertically at a depth of 170 mm +-20 mm. During installation, the ground cover must not be reduced and the roughness of the surface must not be increased. In the present work, a device is presented, that cuts the turf, opens a furrow, incorporates the nonwoven and then closes the furrow. The device is mounted on the tractor and consists of a height guide, a cutting disc, a box coulter and a pressure roller. The device was tested on three plots with a layer structure in Stuttgart. The cutting disc works properly because no clogging was observed. A measurement frame equipped with an ultrasonic sensor, a laser range finder and a feeler wheel determined the surface roughness before and after incorporating the nonwoven. The results showed a significant increase in roughness. In order to reduce the negative impact to the ground surface, it would be possible to increase the ballasting of the device. However, harmful soil compaction must be avoided. The uniform working depth of the developed device was determined with a tachymeter and showed a deviation from the nominal depth of less than 20 mm. The results show that this meets the requirements for the device. Ground cover was measured before and after installing the nonwoven. The turf damage was less than 15 % of the ground cover, which meets the playability requirements. Reconsolidation was determined by penetrologger and evaluated in profile. The soil recompaction, measured as penetration resistance, was similar to the status quo, except in the area close to the nonwoven, where the recompaction failed. The furrows of the developed device can be recompacted more effectively by using two pressure roller, attached V-shaped. But it must be verified that the two pressure rollers do not cause ridge formation, as is the case with seed drills. Overall, the performance of the device can be considered positive, but improvements are still needed to improve reconsolidation. These improvements can be verified in future investigations. At the same time, the effectiveness of the nonwoven must be evaluated in the future. Preliminary tests have shown that the capillary action is sufficient to transport water from the drainage layer to the root zone.
Crop plant reconstruction and feature extraction based on 3-D vision
(2019) Vázquez Arellano, Manuel; Griepentrog, Hans
3-D imaging is increasingly affordable and offers new possibilities for a more efficient agricul-tural practice with the use of highly advances technological devices. Some reasons contrib-uting to this possibility include the continuous increase in computer processing power, the de-crease in cost and size of electronics, the increase in solid state illumination efficiency and the need for greater knowledge and care of the individual crops. The implementation of 3-D im-aging systems in agriculture is impeded by the economic justification of using expensive de-vices for producing relative low-cost seasonal products. However, this may no longer be true since low-cost 3-D sensors, such as the one used in this work, with advance technical capabili-ties are already available. The aim of this cumulative dissertation was to develop new methodologies to reconstruct the 3-D shape of agricultural environment in order to recognized and quantitatively describe struc-tures, in this case: maize plants, for agricultural applications such as plant breeding and preci-sion farming. To fulfil this aim a comprehensive review of the 3-D imaging systems in agricul-tural applications was done to select a sensor that was affordable and has not been fully inves-tigated in agricultural environments. A low-cost TOF sensor was selected to obtain 3-D data of maize plants and a new adaptive methodology was proposed for point cloud rigid registra-tion and stitching. The resulting maize 3-D point clouds were highly dense and generated in a cost-effective manner. The validation of the methodology showed that the plants were recon-structed with high accuracies and the qualitative analysis showed the visual variability of the plants depending on the 3-D perspective view. The generated point cloud was used to obtain information about the plant parameters (stem position and plant height) in order to quantita-tively describe the plant. The resulting plant stem positions were estimated with an average mean error and standard deviation of 27 mm and 14 mm, respectively. Additionally, meaning-ful information about the plant height profile was also provided, with an average overall mean error of 8.7 mm. Since the maize plants considered in this research were highly heterogeneous in height, some of them had folded leaves and were planted with standard deviations that emulate the real performance of a seeder; it can be said that the experimental maize setup was a difficult scenario. Therefore, a better performance, for both, plant stem position and height estimation could be expected for a maize field in better conditions. Finally, having a 3-D re-construction of the maize plants using a cost-effective sensor, mounted on a small electric-motor-driven robotic platform, means that the cost (either economic, energetic or time) of gen-erating every point in the point cloud is greatly reduced compared with previous researches.
Entwicklung einer kontextbasierten Systemarchitektur zur Verbesserung des kooperativen Einsatzes mobiler Arbeitsmaschinen
(2018) Steckel, Thilo; Griepentrog, Hans
In contrast to industrial production processes, agricultural processes are characterized by high uncertainty in terms of planning and execution. Main reasons for this are system-induced high environmental exposure, high complexity of the technical systems, a low division of labor and the lack of applicable systems for decision-making and support.Low process transparency and suboptimal decisions result from that. This observation becomes measurable by comparison of installed performance, determined under ideal conditions, and realized performance, determined from literature and telematics data, which are at a level of approximately 40 to 50%. In the present work the causes for this gap are analyzed and a method for their reduction is developed. Key to improving the situation is the systematic use of context-oriented approaches. The context dimensions time, space and system are described and related to each other. In this way, decision-relevant process conditions in agricultural work processes can be described in a structured manner. On the basis of these contexts, components are derived which, in a subsequent system architecture, enable the automated identification, description and evaluation of process contexts. On this basis concrete measures for the improvement of processes can be derived within the scope of the given possibilities (eg machine performance, drivability). The principle of system architecture is exemplified by the example of the harvesting of silo maize (chipping, transport, storage). The process is modeled from a contextual viewpoint and implemented as agent-based simulation, taking into account the above defined components. In order to carry out the simulation, performance (eg. throughput) and cost-relevant (eg. fuel consumption) parameters are recorded on real machines and production functions are developed. The simulation provides the costs and time requirements for a given process configuration (performance of the forage harvester, number, speed and capacity of the transport vehicles as well as number and mass compacting vehicles). In a parameter configuration based on this simulation a solution space is created which can be used to derive advantageous behaviors. Performance-determining parameters in the determined limits and defines step size are used for that. In addition to the simulation, a mathematical method for the generation of logistic characteristics is developed. Simulation and characteristic fields provide the possibility discreet or continuous approaches of the processes. For verification, results are compared with an empirical survey by questioning farmers and contractors. The described approach allows qualified decisions to improve cooperation in processes and thus contribute to the reduction of the abovementioned performance gap. However, the limits of the improvements result from the locally prevailing environmental conditions and must be defined by the user. Further steps for the control and optimization of processes can be developed on the described approach.
Integrated technical approach for differentiated nitrogen application based on expert knowledge and multiple parameters
(2023) Heiß, Andreas; Griepentrog, Hans
Variable rate nitrogen (N) application is subject to spatio-temporal dynamics of multiple parameters and a high dependency on specific local conditions. Furthermore, existing algorithms are barely capable of considering agronomic expert knowledge and common application technology limits the precise in-field realization. This work approached the complexity of site-specific N management in terms of the decision making, as well as the technical and organizational realization in a systemic manner. A commercial real-time N-sensor system’s behavior was transferred into a fuzzy expert system and extended with soil information. The incorporation into a real-time control included also the spatial synchronization of dose rate determination and realization. A digital process chain to facilitate decision making, data management and execution in the field was conceptualized and evaluated with a prototypical implementation. The N-sensor’s algorithms were precisely imitated with a maximum percentage root mean square error of 0.14%, while the multi-parametric system has implied more robust decisions. In field tests, the real-time control has shown acceptable synchronization errors largely below 1 m and with medians in the range of 0.25 m under realistic conditions. The integrated system architecture has shown a high consistency in terms of straightforward and situative expert knowledge acquisition, as well as the suitability for different sensor and application technologies. The work represents a systemic approach for a derivation and employment of machine-readable algorithms from agronomic expert knowledge defining the cause-effect relationships for a site-specific N application. Its generic properties allow a supplementation by other models and can in turn strengthen them further.
Modelling and optimisation of no-till seeder dynamics for precise seeding depth
(2019) Sharipov, Galibjon; Griepentrog, Hans
Achieving better seeding depth consistency in no-till seeding is a critical performance metric of the seeding machine and is of great importance due to its profound effect on reliable seed germination and seedling emergence resulting in a yield increase. Growing implementation of no-tillage in big size farms requires high-capacity seeding machines with increased operation speed and working width. Thus, the increased capacity of the seeding machine as well as harsh soil conditions like the surface undulations and the presence of previous crop residues make the desired working quality of no-till seeders challenging for both designers and manufacturers. The aim of this cumulative dissertation was to optimise a no-till seeder dynamics in terms of vertical motion stability for better seed placement under realistic high-capacity performance. To fulfil this aim, an approach to achieve the desired dynamic behaviour of the seeder was carried out based on three phases: (1) evaluation of the seeder dynamic performance by defining the relationship between the seeder dynamics and the corresponding seeding depth variation, (2) modelling and simulation of the seeding assembly motion dynamics to specify a control system (e.g. MR damper system) for dynamics improvement, (3) implementation of the defined system into the seeding assembly and testing of the new seeding assembly prototype. The present work was the first approach to optimise the dynamic motion behaviour of a no-till seeder by implementing an MR damper system into its seeding assembly for better seed placement under realistic high-capacity working conditions. The AMAZONEN no-till direct seeder was an ideal candidate for this investigation as it contains 12 identical tine type seeding assemblies where the operating depth is defined by the position of the packer wheel. Under working conditions, the maximum width is 3 m resulted from the inter-row distance of 0.25 m between the seeding assemblies. The seeding assemblies are provided with downforces using a hydraulic cylinder in order to keep the packer wheel of the assemblies on the ground and to maintain a consistent seeding depth during seeding operation. Concurrent and geo-referenced sensor data made it possible to acquire the dynamics parameters of the seeder and the corresponding soil surface profiles (the point where the packer wheel touches the ground). This together with the measured 3D geo-referenced position of the seeds gave the opportunity to define the reason of high variations in seeding depth. A sensor-frame was developed, utilising up-to-date sensor technology, to capture the seeder dynamics and to determine the corresponding soil surface profile. A combination of strains recorded at the three corresponding points of the seeding assembly using linear strain gauges was employed to calculate the vertical forces, draught forces and the profile impact forces. A new methodology was introduced to extract the absolute seeding depth from the combination of the determined surface profile and the measured 3D position of the seeds in absolute coordinates. Geo-referenced coordinates of seed positions in combination with geo-referenced surface profile and machine dynamics parameters, offered the possibility to define the reason of seeding depth variation. To do that, the relation between the forces (i.e. vertical and profile impact forces) and the variation of seeding depth was defined by correlating the spatial frequency contents of each dataset. An investigation of the seeder dynamics was carried out by modelling and simulating its performance based on measured data (e.g. determined surface profile and vertical forces) to define a system that can reduce the effect of the forces for better seed placement in no-till seeding. The seeding assembly together with and without a MR (magnetorheological) damper system, which was considered to be located in-between the coulter and the packer wheel, was introduced as a semi-active and passive system. Furthermore, three hysteresis models, such as Bingham, Dahl and Bounc-Wen model, were applied for the semi-active MR damper system behaviour. Among the models, the Bouc-Wen model demonstrated more significant improvements over the passive system model. Analysis of the performance of the semi-active MR damper implemented seeding assembly against the passive system proved the vertical motion dynamics of the assembly, in terms of vertical displacements (52.3%) and its affecting forces (54.1%) to be optimised for better seed placement. Testing the performance of the MR damper implemented seeding assembly compared with that of the original seeding assembly confirmed the potential of the MR damper implemented seeding assembly. The dynamics of the seeding assembly with the MR damper depicted a reduction of 67.69% in the amplitude of the impact forces compared to the original seeding assembly. Consequently, the improvement in the dynamics resulted in better seed placement. The variation of the damped seeding depth, as it was the performance of the seeding assembly with the MR damper, compared to the target seeding depth resulted in an absolute error of 11.9 mm for 95% of its samples, which is considerably less than the error with a value of 21.3 mm for the seeding depth variation resulted from the original seeding assembly. By designing the seeding assembly with the MR damper system, the dynamics of seeding machine can be significantly optimized for better seeding depth consistency.
Perception for context awareness of agricultural robots
(2018) Reiser, David; Griepentrog, Hans
Context awareness is one key point for the realisation of robust autonomous systems in unstructured environments like agriculture. Robots need a precise description of their environment so that tasks could be planned and executed correctly. When using a robot system in a controlled, not changing environment, the programmer maybe could model all possible circumstances to get the system reliable. However, the situation gets more complex when the environment and the objects are changing their shape, position or behaviour. Perception for context awareness in agriculture means to detect and classify objects of interest in the environment correctly and react to them. The aim of this cumulative dissertation was to apply different strategies to increase context awareness with perception in mobile robots in agriculture. The objectives of this thesis were to address five aspects of environment perception: (I) test static local sensor communication with a mobile vehicle, (II) detect unstructured objects in a controlled environment, (III) describe the influence of growth stage to algorithm outcomes, (IV) use the gained sensor information to detect single plants and (V) improve the robustness of algorithms under noisy conditions. First, the communication between a static Wireless Sensor Network and a mobile robot was investigated. The wireless sensor nodes were able to send local data from sensors attached to the systems. The sensors were placed in a vineyard and the robot followed automatically the row structure to receive the data. It was possible to localize the single nodes just with the exact robot position and the attenuation model of the received signal strength with triangulation. The precision was 0.6 m and more precise than a provided differential global navigation satellite system signal. The second research area focused on the detection of unstructured objects in point clouds. Therefore, a low-cost sonar sensor was attached to a 3D-frame with millimetre level accuracy to exactly localize the sensor position. With the sensor position and the sensor reading, a 3D point cloud was created. In the workspace, 10 individual plant species were placed. They could be detected automatically with an accuracy of 2.7 cm. An attached valve was able to spray these specific plant positions, which resulted in a liquid saving of 72%, compared to a conventional spraying method, covering the whole crop row area. As plants are dynamic objects, the third objective of describing the plant growth with adequate sensor data, was important to characterise the unstructured agriculture domain. For revering and testing algorithms to the same data, maize rows were planted in a greenhouse. The exact positions of all plants were measured with a total station. Then a robot vehicle was guided through the crop rows and the data of attached sensors were recorded. With the help of the total station, it was possible to track down the vehicle position and to refer all data to the same coordinate frame. The data recording was performed over 7 times over a period of 6 weeks. This created datasets could afterwards be used to assess different algorithms and to test them against different growth changes of the plants. It could be shown that a basic RANSAC line following algorithm could not perform correctly under all growth stages without additional filtering. The fourth paper used this created datasets to search for single plants with a sensor normally used for obstacle avoidance. One tilted laser scanner was used with the exact robot position to create 3D point clouds, where two different methods for single plant detection were applied. Both methods used the spacing to detect single plants. The second method used the fixed plant spacing and row beginning, to resolve the plant positions iteratively. The first method reached detection rates of 73.7% and a root mean square error of 3.6 cm. The iterative second method reached a detection rate of 100% with an accuracy of 2.6 - 3.0 cm. For assessing the robustness of the plant detection, an algorithm was used to detect the plant positions in six different growth stages of the given datasets. A graph-cut based algorithm was used, what improved the results for single plant detection. As the algorithm was not sensitive against overlaying and noisy point clouds, a detection rate of 100% was realised, with an accuracy for the estimated height of the plants with 1.55 cm. The stem position was resolved with an accuracy of 2.05 cm. This thesis showed up different methods of perception for context awareness, which could help to improve the robustness of robots in agriculture. When the objects in the environment are known, it could be possible to react and interact smarter with the environment as it is the case in agricultural robotics. Especially the detection of single plants before the robot reaches them could help to improve the navigation and interaction of agricultural robots.